Lower Mara River Environmental Flow Assessment Final Resource Quality Objectives and Reserve Assessment Report NELSAP Technical Reports: Basin Development Series 2020 - 01 APRIL 2020 Biodiversity Conservation and Sustainable Utilization of Ecosystem Services of Wetlands of Transboundary Relevance in the Nile Basin Determination of environmental flow requirements for the lower Mara as part of the Transboundary Water Allocation Planning in the Lower Mara River Basin (Tanzania)
Prepared by the Nile Basin Initiative Nile Equatorial Lakes Subsidiary Action Program Coordination Unit (NELSAP-CU) with financial support from Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), on behalf of the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) under the International Climate Initiative for the Ministry of Water of the United Republic of Tanzania under technical support from IHE Delft Institute for Water Education.
April 2020
The purpose of the technical report series is to support informed stakeholder dialogue and decision making in order to achieve sustainable socio-economic development through equitable utilization of, and benefit from, the shared Nile Basin water resources.
Funding source: German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) under the International Climate Initiative Project title: Conserving biodiversity in the Nile Basin transboundary wetlands Project number: 14.9029.1
Disclaimer The views expressed in this publication are not necessarily those of NBI’s Member States or its development partners. Trademark names and symbols are used in an editorial fashion and no intention of infringement on trade mark or copyright laws. While every care has been exercised in compiling and publishing the information and data contained in this document, the NBI regrets any errors or omissions that may have been unwittingly made in this publication. The NBI is not an authority on International Administrative Boundaries. All country boundaries used in this publication are based on FAO Global Administrative Unit Layers (GAUL).
© 2020 Nile Equatorial Lakes Subsidiary Action Program (NELSAP-CU) / Nile Basin Initiative (NBI)
2 A SECTION OF THE LOWER MARA RIVER IN TANZANIA
EXECUTIVE SUMMARY
Introduction and Study Area This report presents the resource quality objectives (RQOs) and a reserve assessment for the Lower Mara River Basin (Lower MRB) of Tanzania. Activities reported are part of the project “Biodiversity Conservation and Sustainable Utilization of Ecosystem Services of Wetlands of Transboundary Relevance in the Nile Basin”, which seeks to build consensus and enhance cooperation in water resources management and development between the Nile Basin’s riparian countries. Results also contribute to the development of a water allocation plan for the Lower MRB under the Memorandum of Understanding signed by Tanzania and Kenya in 2015 for Joint Water Resources Management of the Transboundary Mara River Basin. The environmental Flow assessment report was developed and in close cooperation with and direct involvement of the Lake Victoria Basin Water Board (LVBWB) under of the Ministry of Water of Tanzania, which is the authority legally responsible for setting and protecting RQOs and the reserve in water resource management. The process involved Resource Quality Objective (RQO) setting through a stakeholder’s workshop, Environmental flow assessment and Reserve flow setting by specialists and water authorities. RQOs and the reserve are a requirement in Part VI (Protection of Water Resources) of the 2009 National Water Resources Management Act of Tanzania. RQOs are intended to protect water and related aquatic biological resources at levels needed to meet the needs of resource users and maintain ecosystems in a desired environmental management class. The reserve is defined as the quantity and quality of water required for; (a) satisfying basic human needs by securing a basic water supply for people who are now or who shall in the reasonably for near future, be; (i) relying upon (ii) taking water from; or (iii) being supplied from the relevant water resources; and (b) protecting aquatic ecosystems in order to secure ecologically sustainable development and use of the relevant water resources. The reserve contributes to achieving the RQOs and consists of two parts. Part 1 is focused on meeting basic human needs, which can be considered a component of the domestic water demand, and part 2 is focused on protecting aquatic ecosystems. In this study we distinguish between the basic human needs component of the reserve and the ecological component of the reserve, known as environmental flows. Environmental flows are defined in the NBI’s Strategy for Management of Environmental Flows in the Nile as the quantity, timing and quality of water flows required to sustain freshwater and estuarine ecosystems and the human livelihoods and well-being that depend upon these ecosystems. Setting RQOs and the reserve in the Lower MRB is a priority because of the need to balance the growing water needs of the population with the conservation of world-class ecosystems of the basin. The 2018
3 population of the Lower MRB is estimated at 396,000 and is projected to grow to more than 700,000 by 2030. To develop sustainably, this population must share the basin’s limited water resources with ecosystems of Serengeti National Park, the Mara Wetland, and aquatic ecosystems extending up into each of the Mara River’s tributaries. Protecting aquatic ecosystems is not only a requirement of the law, it is also important to people’s health and livelihoods as communities rely on these ecosystems for many services, including i) clean water for people, agriculture, and livestock, ii) food in the form of fish and edible wild plants, iii) herbal medicines and other natural products, and iv) religious and cultural services central to the identities of the
communities. All of these factors are taken into account in setting RQOs for the Lower MRB and assessing the reserve flows needed to meet them. FIGURE ES 1: MAP OF THE MARA RIVER BASIN
Process for Setting RQOs and the Reserve The determination of the reserve and the related RQOs was completed using the NBI Environmental Flows Framework (NBI, 2016a, 2016b). This framework was developed by NBI to ensure a standard process is followed for the increasing number of environmental flow assessments being conducted in the Nile Basin. There are seven main steps in this framework, which are summarized in Figure ES-2.
Phase 1 included a policy review, compiled available information related to environmental flows (including field data, scientific literature, project reports, and other environmental flow assessments completed in the Mara River Basin and Tanzania), integrated our efforts with on-going water resources related work in the basin, and strengthened partnerships with relevant government organizations, projects, and non-government organizations, with a focus on partnerships and capacity building within the LVBWB and the Ministry of Water. Phase 2 initiated the process of setting RQOs for individual resource units (Figure ES-3) by assembling a group of stakeholders to contribute to the process and holding a workshop to gather their inputs. These efforts produced assessments of pressures, important activities, and conditions in each resource unit, preferred management class for each resource unit, and draft RQO statements for each resource unit for quantity, quality, habitat, and biota. The involvement of the Water User Associations was central to this effort. 4 FIGURE ES 2: SUMMARY OF THE STEPS INVOLVED IN THE NBI E-FLOWS FRAMEWORK (NBI, 2016B)
Phase 3 analysed the hydrological foundation by regionalizing available datasets to quantify monthly and average annual discharge, minimum and maximum discharge values, annual and monthly flow duration curves, and maximum daily flow frequency analysis. Phase 4 reviewed existing ecosystem and river classification systems or maps used in Tanzania and conducted ecological, biological, and geomorphology assessments to determine the characteristics of study sites selected for more detailed investigation. This resulted in the classification of areas by mainsteam, tributary, and wetland features. Phase 5 evaluated the potential deviation of current-condition flows from baseline- (or natural-) condition flows. Significant flow alterations occur in river basins regulated by large infrastructure orwithhigh water demand relative to water availability. In the Mara, there are currently no engineered structures that significantly alter flows. Thus, under most flow conditions the flow regime is near natural. Phase 6 implemented a modified Building Block Methodology to assess flow-ecology ecosystem services linkages. This involved detailed analyses by a team of specialists in hydrology, hydraulics, water quality, geomorphology, fish, macroinvertebrates, riparian vegetation, and social use. Social surveys were carried out in 14 villages and biophysical surveys were conducted at seven sites, including two on the mainstem Mara River, three near the mouths of major tributaries, and two in the Mara Wetland (Figure ES-4). Biophysical surveys were conducted during two time periods. Reporting of the findings from this work forms the bulk of this document. Phase 7 consisted of a flow setting technical meeting to synthesize the results of the team of specialists and
5 FIGURE ES-3: RESOURCE UNITS FOR THE LOWER MRB set values for the ecological component of the reserve that met RQOs. It also included quantification of the basic human needs component of the reserve and recommended monitoring activities for compliance and effectiveness monitoring. Results RQOs set during the project reflect the close and multifaceted interdependencies of people and water and aquatic ecological resources in the Mara River Basin of Tanzania. People depend on river flows to meet water needs for domestic purposes, livestock, and agriculture across the basin. Groundwater is also an important source for domestic water. Special emphasis was given to dry season flows, but the importance of wet season flows was also highlighted for supporting floodplain agriculture and replenishing surface and groundwater storage for use in subsequent dry seasons. The importance of ecosystem processes is recognized as maintaining an ambient level of water quality needed for healthy fisheries and water for domestic uses, livestock, and agriculture. Instream and riparian habitats and related biota are valued for the direct resources they provide (fish, building materials) as well as their role in supporting biodiversity. Biodiversity protection is recognized as the predominate use for water in Serengeti National Park but was also noted as important to inhabitants in all parts of the basin. These dependencies and values are recorded in the RQOs set by stakeholders. In all resource units (aligned with sub-basins) of the Mara, objectives were set to maintain ecosystems in no less than a somewhat altered condition, which corresponds to a class of B in the draft River Classification System for Tanzania. In this class, the “natural flow regime is affected by water withdrawals, impoundments and/or discharges, but the critical aspects of the flow regime are retained so that effects on the ecosystem are small.” The final estimates for flow requirements for basic human needs were calculated in units of m3/day and also m3/s to align with the environmental flow values. The basic human need values are approximately based on
6 FIGURE ES 4: MAP OF THE STUDY SITES FOR THE SOCIAL SURVEY, BIOPHYSICAL FIELD CAMPAIGNS, AND RUS
the resource units to align with the planning units used in the water allocation planning effort. Upper and Lower Tigithe resource units have been combined as “Tigithe” and North Mara and South Mara RUs as “Mara Wetland”, for a total of 6 resource units. The basic human need values for 2018 ranged from 0.009 m3/s in
Tobora resource unit to 0.039 m3/s in the Mara Wetland resource unit (Table ES-1). The values for basic human needs are based on a daily requirement of 25 liters/person/day and are expected to remain constant throughout the year. Monthly low flows for the ecological component of the reserve at riverine sites are presented relative to average flows in Figure ES-5. Flows, or depth in the case of the Mara Wetland site, during the driest and wettest
7 months are presented in Table ES-2 (maintenance year) and Table ES-3 (drought year). Also presented are high flows characterized by a magnitude, duration, and timing. During years of normal rainfall, results for the environmental flow of mainstem Mara River sites corresponded to 28 percent of the value of the average flow of the wettest month (May) and 22 to 27 percent of the average flow of the driest month (August). Flows in excess of the environmental flow are expect at least 95 percent of the time based on the available hydrological data. Environmental flows determined for mainstem sites during drought years were roughly 33 percent lower than those for normal years. Environmental flows for tributaries and the wetland correspond to larger or smaller proportions of estimated average monthly flow and may even exceed the monthly average during the driest month. The more extreme proportions at these sites are predominantly due to uncertainties in the estimation of hydrological regimes, which were regionalized from the relationships between precipitation data and mainstem hydrological records. Environmental flows for normal years are intended to support the full range of ecological processes needed to
FIGURE ES 5: MONTHLY LOW FLOW REQUIREMENTS OF THE ECOLOGICAL COMPONENT OF THE RESERVE FOR RIVERINE STUDY SITES. BECAUSE OF EXPECTED NO-FLOW CONDITIONS DURING DROUGHT IN THE TOBORA AND SOMOCHE TRIBUTARIES, DROUGHT FLOW LEVELS WERE NOT SET FOR THESE SITES. 8 9 maintain healthy plant and animal communities in the river system. Protection of ecological processes in the river also ensured continued delivery of ecosystem services beneficial to human communities. Environmental flows for drought years are intended to sustain life in the system until higher flow levels return.
Conclusions The RQOs and reserve levels determined in this project comply with all requirements and approved guidelines under Tanzanian law and thus qualify for notice in the Gazette and application in continued water resource planning. They are judged to be valid for a period of five years, which corresponds to the validity period of the basin Integrated Water Resource Management and Development Plan. After this period, and in the context of regular water resource planning and management, they should be reviewed and revised if judged necessary. Both the RQOs and reserve flows set in this assessment are relevant for the water allocation plan currently under development by the LVBWB and the Ministry of Water. According to the draft guidelines for water allocation planning developed by the Ministry of Water, water resources allocation is “a means by which regulation of water use is done through sharing water resources among competing users, with due regard for the environment, the economy, and the social wellbeing of all Tanzanians”. Setting RQOs is a required step in this process and a mechanism to incorporate stakeholder interests and align them with the requirements of Tanzanian laws and regulations. The reserve is a term required in the water balance to be calculated for each water allocation planning unit, represented by the equation: Water Balance = Available Water – (Reserve + Transfers + Summation of Water Allocations) A positive water balance indicates that there is sufficient available water to meet all water demands, while a negative balance indicates a state of over allocation. The water balance can be calculated at monthly, seasonal, or annual time intervals. The results of this assessment quantified both the basic-human-need and ecological components of the reserve at monthly intervals, which allows them to be incorporated into the water balance at whatever time interval is chosen. It will still be necessary, however, to extrapolate reserve values to the outlets of final planning units. This can be done by adjusting the values reported in this document in proportion to the upstream contributing area of each planning unit. The reserve values are also relevant to the management stage of water allocation planning, including aspects of compliance and enforcement. In the implementation of the water allocation plan, river levels are to be monitored to determine whether water users may continue to withdrawal water at the full limit of their permit or whether restrictions should be imposed to protect reserve flows in the river. Current estimates are that 75 percent of the water flowing in the Mara River in Tanzania comes from Kenya. Thus, close coordination is necessary between the countries in water allocation and management. This also applies to consideration of the reserve. Fortunately, Tanzanian and Kenyan water laws are consistent in their definition of the reserve and assigning it highest priority in water allocation. Both countries include basic human needs and ecosystem protection as components of the reserve. Both countries recognize the basic human need to be 25 liters/person/day, and both countries have adopted the Nile E-Flows Framework for the determination of the ecological component for transboundary rivers. This consistency in laws, definitions, and approaches greatly enhances the potential for harmonious management of water resources across the border. Care must also be taken that numerical values of reserve flows and implementation measures are consistent in a manner that ensures Kenyan reserve flows crossing the border are sufficient to meet Tanzanian reserve flow levels. The environmental management objectives of Tanzania and Kenya at the border are similar given the juxtaposition of Serengeti National Park and Maasai Mara National Reserve. This should lead to similar determinations of the ecological component of the reserve. The reserve determined in this assessment at Kogatende in Serengeti National Park is judged sufficient to meet downstream reserve requirements in the 5-year time period these determinations will remain valid. Knowledge Gaps Uncertainty is inevitable in any scientific assessment of reserve levels, especially in data scarce systems like the Mara River Basin. This assessment has been transparent in acknowledging uncertainties and taking steps to minimize risks associated with them. The assessment team stands behind the reserve flows reported here but also strongly recommends that actions be taken to improve knowledge and understanding of key components of the resource system. Urgent action is needed to restore the hydrometeorological monitoring network of the Mara River Basin. There are currently no functional river discharge or precipitation stations in the basin. In this assessment, suitable historical data were available from only one river discharge station (Mara Mines) and two precipitation stations (Nyabassi and Mugumu) which are near but outside the basin. Almost nothing is known about groundwater, which was not explicitly considered in the reserve assessment. The lack of historical hydrological data had a minimal impact on reserve flows determined in this assessment because the modified building block method
10 used is based primarily on data collected during the assessment itself. Daily precipitation and flow data are necessary for proper implementation of the reserve, and long-term data sets are necessary for broader planning of water resource use and allocation. The lack of long-term hydrological data for the Somoche, Tobora, and Tigithe Rivers is of concern for water allocation because of the high uncertainties associated with the regionalized data from the water resource assessment. This leads to uncertainty in the total quantity of water available during different months of the year and between different years. So, while there is higher confidence in the reserve flows, the uncertainty in the total water available is transferred to the water balance and volume of water available for allocation to uses like domestic, livestock, irrigation, and industry. If regionalized data overestimate the total water available this could lead to over-allocation of water in permits. The hydrology and hydraulics of the Mara Wetland also remain largely unknown. During the field assessments the team measured flows in the wetland that significantly exceeded flows into the wetland at Mara Mines. This indicates that flows in the wetland included drainage of stored water as well as inflows from the Mara River. Water levels in the lower portions of the wetland also appear to be influenced by the level of Lake Victoria, which diminishes the degree to which these portions of the wetland are dependent on Mara River flows. Improved knowledge of these hydraulic characteristics of the wetland, bathymetry of the wetland, and associated plant and animal communities is needed to set appropriate reserve levels in Mara River. Finally, because the Mara River is presently most vulnerable to flow alterations under low flow conditions, there is urgency to improve knowledge of how aquatic ecosystems function during low flows. Low flows are a natural part of the river’s hydrograph and riverine and wetland species are adapted to cope with natural
11 low flow conditions. But the increasing water demands of basin inhabitants during dry periods are likely to reduce river flows to unnatural levels and to extend the duration of low flows. This will increase stress on river organisms to levels beyond their adapted tolerance levels.
Contributors
Prof. Michael McClain, Project Leader, IHE Delft Institute for Water Education, Delft, the Netherlands Lauren Zielinski, Project Coordinator, IHE Delft Institute for Water Education, Delft, the Netherlands Anne Siema, Water Quality, Nowell Consultants, Kenya Prof. Felister Mombo, Socioeconomics, Sokoine University of Agriculture, Tanzania Dr. Gordon O’Brien, Fish Ecologist, University of Mpumalanga, South Africa Dr. Jochen Wenninger, Hydrologist, IHE Delft Institute for Water Education, Delft, the Netherlands Dr. Bennie van der Waal, Geomorphology & Hydraulics, Rhodes University, South Africa Dr. Frank Masese, Macroinvertebrates, University of Eldoret, Kenya Dr. Samora Macrice, Riparian Vegetation, Sokoine University of Agriculture, Tanzania Sadiki Lotha Laiser, Biodiversity Conservation Expert, Nile Basin Initiative Mwita Mataru, Lake Victoria Basin Water Board, Tanzania Wiston Robert, Lake Victoria Basin Water Board, Tanzania John Ngawambala, Lake Victoria Basin Water Board, Tanzania Qambemeda Nyanghura, Socioeconomics, Sokoine University of Agriculture, Tanzania David Bigirwa, Socioeconomics, Sokoine University of Agriculture, Tanzania Andrew Husted, Fish Ecologist, The Biodiversity Company, South Africa Christian Fry, Fish Ecologist, The Biodiversity Company, South Africa This report was unanimously approved by the Water Allocation Planning Technical Writing team and endorsed as a constituent to the Water Allocation Plan after the inclusion of comments gathered in the 20 September 2019 validation workshop attended by: Dr. George Lugomela, Director of Water, Ministry of Water Sylvester Matemu, Assistant Director, Transboundary Water Resources, Ministry of Water Tumaini W.M Mwamyalla, Sr. Community Development Officer, Nile Basin Initiative/Ministry of Water David Manyama, Ministry of Water Modestus Zacharia, Ministry of Water Fredrick M. Mgube, Lake Victoria Basin Commission John B. Ngawambala, Community Development Officer, Lake Victoria Basin Water Board - Musoma Ogoma Mangasa, Hydrologist, Lake Victoria Basin Water Board – Mwanza Isabela Tenganiza, Lake Victoria Basin Water Board – Mwanza Gordon Mumbo, Chief of Party, Sustainable Water Partnership Clare Haule, Project officer, Sustainable Water Parternship - Tanzania Christian Chonya, Manager, Fresh Water, World Wide Fund for Nature - Tanzania 12 Maro Andy Tola, Program Officer, WRM&D, Nile Equatorial Lakes Subsidiary Action Program Sadiki Lotha Laiser, Program Officer, Wetlands, Nile Equatorial Lakes Subsidiary Action Program Odette Omutega, Nile Equatorial Lakes Subsidiary Action Program Acknowledgements
Development of this Environmental Flow Report is made possible through technical support from UNESCO- IHE Delft through its consultacy services. Special thanks are to the Federal Government of Germany, Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) for their financial support and Nile Basin Initiative (NBI) Nile Equatorial Lakes Subsidiary Action Program Coordination Unit (NELSAP CU) and through Gesellschaft für Internationale Zusammenarbeit (GIZ) for their Strategic and programmatic support of the project given by Juan Carlos Sanchez, Advisor for GIZ, and Sadiki Lotha Laiser, Biodiversity Conservation and Sustainable Utilization of Ecosystem Services/wetlands Expert for NBI. The Initiative extends its thanks to Lake Victoria Basin Water Board (LVBWB) for engaging fully in the project from day one, including providing institutional knowledge on the water resources management in the Lower Mara River Basin, planning and logistical support on field work and workshops, connecting the team with the right partner agencies, and providing guidance on local customs to make the project a success. The technical expertise and hard work of Eng. Florence Mahay, Lake Victoria Basin Water Board Officer, Mwita Mataru, Environmental Expert, John Ngawambala, Community Development Officer, Wiston Robert Agwaro, Hydrologist, Ogoma Mangasa Nyamhanga, Hydrologist, and Gerald Itimbula, Communications Officer, were critical for the success for the project. The project team would also like to thank the LVBWB drivers for their dedication transporting the field team and their equipment across the basin: Friday Mwanizumbha, William Sanda, Amosi Kishebuka, and Constantine Mirumbe. The Water Allocation Planning task force members from the Ministry of Water in Tanzania for their active participation, policy guidance, and review of technical documents: Modestus Herman Mballa, Estella Mgala, David Manyama Jamberi, Ramadhani Hamza Singano, and Hosea Sanga. The entire EFA technical team and project assistants for their strong dedication to the project and providing valuable contributions to all aspects of the project, including Anne Siema, Dr. Gordon O’Brien, Dr. Jochen Wenninger, Dr. Bennie van der Waal, Dr. Frank Masese, Dr. Samora Macrice, Dr. Felister Mombo, Patrick Meya Wadheir, Qambemeda Nyanghura, David Bigirwa, Christian Fry, Andrew Husted, and Elizabeth Grater. Partners agencies who assisted during the field campaigns and workshops, including Kelvin Mollel, Park Ecologist, and William Mwakilema, Chief Park Warden, from Serengeti National Park; Jonas Nyandwi, Fisheries Officer, Mlela Noah, Assistant Fisheries Officer, and Mohamed Ngauna, Rajabu Hegga, and Gosbeth Buchard, boat engineers and drivers, from the Musoma District Fisheries Department for their generous assistance with fisheries equipment and expertise; and the Water Resources Authority in Kenya for their continued cooperation in the transboundary management of the Mara River Basin as well as providing technical staff and equipment vital to the project. Long-standing project partners, USAID’s Sustainable Water Partnership and WWF Tanzania for their continued contributions and support of this project and promoting sustainable water management in the Mara River Basin.
13 14 The Initiative offer thanks to the many members of local government and the community who assisted our field campaigns, including water users’ associations, district water and environment officers, and local villagers who provided critical local knowledge and insight to the project. Table of Contents Executive Summary i ES.1 Introduction and Study Area i ES.2 Process for Setting RQOs and the Reserve ii ES.3 Results iv ES.4 Conclusions x ES.5 Knowledge Gaps xi 1 Introduction 1-1 1.1 Project Scope and Objectives 1-1 1.2 Resource Quality Objectives, Environmental Flows and the Reserve 1-2 1.3 Cooperation with National and Transboundary Water Allocation Planning Activities 1-3 1.4 Content of this Report 1-3 2 Overview of the Lower Mara River Basin 2-1 2.1 Administration 2-2 2.2 Hydrology 2-3 2.3 Climate 2-4 2.4 Land Use 2-7 2.5 Physiography and Geomorphology 2-8 2.6 Communities and Social Importance 2-10 3 Reserve Assessment Methodology 3-1 3.1 Basin Scale Situation Assessment and Alignment Process 3-5 3.1.1 Tanzania Policy Review 3-5 3.1.2 Available Information 3-10 3.1.3 Previous EFAs in the Mara River Basin 3-10 3.1.4 Water Resources Activities in the Mara River Basin 3-11 3.1.5 Partnerships and Capacity Building 3-12 3.2 Resource Quality Objective Setting 3-13 3.2.1 Resource Units 3-14 3.2.2 Stakeholder Workshop 3-14 3.3 Hydrologic Foundation 3-16 3.3.1 Data Regionalization 3-16 3.3.2 Long-term Annual Water Balance 3-16 3.3.3 Regionalized Monthly and Annual Flow Values 3-17 3.3.4 Frequency Analysis 3-18 3.3.5 Trend Analysis 3-18 3.3.6 Hydrological Assessment of EFA Study Sites 3-19 3.4 Ecosystem Classification 3-19 3.5 Flow Alterations 3-21 3.6 Flow-Ecology Ecosystem Services Linkages 3-21 3.6.1 Flow-Ecology Components 3-23 3.6.2 EFA Study Sites and Field Campaigns 3-29 3.6.3 Starter Documents 3-33 3.6.4 Flow Setting Technical Meeting 3-36 3.7 Reserve Setting and Monitoring 3-37 3.7.1 Reserve Values 3-37 3.7.2 Monitoring and Adaptive Management 3-37 4 Study Site Results 4-1 4.1 Kogatende 4-1 4.1.1 Social Survey 4-2 4.1.2 Biophysical – Mara River at Kogatende 4-2 4.1.3 Site Metrics 4-9 4.1.4 Indicators and Management Objectives 4-10 15 4.1.5 Required Conditions 4-11 4.1.6 Confidence Ratings 4-12 4.2 Tobora 4-12 4.2.1 Social Survey – Matare, Nyamerama 4-13 4.2.2 Biophysical – Tobora River at Nyasurura 4-16 4.2.3 Site Metrics 4-24 4.2.4 Indicators and Management Objectives 4-24 4.2.5 Required Conditions 4-25 4.2.6 Confidence Ratings 4-26 4.3 Somoche 4-27 4.3.1 Social Survey – Kwitete, Nyamitita 4-28 4.3.2 Biophysical – Somoche River at Somoche 4-30 4.3.3 Site Metrics 4-37 4.3.4 Indicators and Management Objectives 4-38 4.3.5 Required Conditions 4-39 4.3.6 Confidence Ratings 4-40 4.4 Tigithe 4-41 4.4.1 Social Survey 4-42 4.4.2 Biophysical – Tigithe River at Matongo 4-46 4.4.3 Site Metrics 4-54 4.4.4 Indicators and Management Objectives 4-54 4.4.5 Required Conditions 4-56 4.4.6 Confidence Ratings 4-57 4.5 Mara Mines 4-57 4.5.1 Social Survey – Borenga, Gantamome 4-58 4.5.2 Biophysical – Mara River at Mara Mines 4-60 4.5.3 Site Metrics 4-69 4.5.4 Indicators and Management Objectives 4-70 4.5.5 Required Conditions 4-71 4.5.6 Confidence Ratings 4-72 4.6 Bisarwi 4-73 4.6.1 Social Survey – Kembwi, Marasibora 4-74 4.6.2 Biophysical – Mara Wetland at Bisarwi 4-76 4.6.3 Site Metrics 4-83 4.6.4 Indicators and Management Objectives 4-83 4.6.5 Required Conditions 4-84 4.6.6 Confidence Ratings 4-85 4.7 Mara Wetland 4-86 4.7.1 Social Survey – Ryamisanga, Wegero 4-87 4.7.2 Biophysical – Mara Wetland at Ketasakwa 4-89 4.7.3 Site Metrics 4-96 4.7.4 Indicators and Management Objectives 4-97 4.7.5 Required Conditions 4-98 4.7.6 Confidence Ratings 4-99 4.8 Data Gaps 4-99 5 Final Resource Quality Objectives and Reserve Recommendations 5-1 5.1 Final Resource Quality Objectives 5-1 5.1.1 Kogatende 5-1 5.1.2 Tobora 5-3 5.1.3 Somoche 5-5 5.1.4 Tigithe 5-7 5.1.5 Mara Mines 5-10 5.1.6 Bisarwi 5-12 5.1.7 Mara Wetland 5-14 5.2 Basic Human Needs 5-16
16 5.3 Environmental Flows 5-18 5.3.1 Kogatende 5-21 5.3.2 Tobora 5-25 5.3.3 Somoche 5-29 5.3.4 Tigithe 5-32 5.3.5 Mara Mines 5-35 5.3.6 Bisarwi 5-38 5.3.7 Mara Wetland 5-41 5.4 Monitoring and Adaptive Management 5-42 5.4.1 Monitoring for the Reserve 5-42 5.4.2 Incorporating Monitoring Data into Adaptive Management 5-49 6 Discussion and Uncertainties 6-1 6.1 Implementation of RQOs and Reserve Flows in Water Allocation Planning 6-2 6.2 Harmonization of Reserve Flows with those set for Kenya 6-3 6.3 Knowledge Gaps to Address 6-4 6.3.1 Basin Hydrology 6-4 6.3.2 Wetland Hydrodynamics 6-5 6.3.3 Low-Flow Ecology 6-5 6.4 Special Considerations 6-5 6.4.1 Climate Change 6-5 6.4.2 Heavy Metal Pollution from Mining Activities 6-7 6.4.3 Wildlife 6-8 6.5 Integration into Existing and Future Water Resources Management 6-9 7 References 7-1 8 Annexes 8-1 Annex A Stakeholder Input on Determining Resource Quality Objectives for the Lower Mara River, Workshop Report Annex B Starter Document for Socioeconomics Annex C Starter Document for Hydrology Annex D Starter Document for Hydraulics Annex E Starter Document for Geomorphology Annex F Starter Document for Riparian Vegetation Annex G Starter Document for Fish Annex H Starter Document for Macroinvertebrates Annex I Starter Document for Water Quality Annex J Flow Setting Technical Meeting Report
List of Figures Figure 2 1: Map of the Mara River Basin 2-1 Figure 2 2: Major river basins located in Tanzania 2-2 Figure 2 3: Districts located in the Lower MRB 2-3 Figure 2 4: HUs contained in the Lower MRB 2-4 Figure 2 5: Spatial distribution of precipitation stations in the Lower MRB and Thiessen polygon areas 2-5 Figure 2 6: Average annual precipitation at stations relevant to the Lower MRB 2-5 Figure 2 7: Average monthly discharge (left) and runoff regimes (right) at HU1 outlets 2-5 Figure 2 8: Average monthly precipitation at stations relevant for the Lower MRB 2-6 Figure 2 9: Land use map of the Mara River Basin (MaMaSe, 2019) 2-7 Figure 2 10: Land use classes within the lower MRB catchments HU1 (left), HU2 (centre), and HU3 (right) 2-8 Figure 2 11: A longitudinal profile of the Mara River showing the EFA sites of 2016 as orange dots and the current biophysical study sites as black dots 2-8 Figure 2 12: Location of the sampling sites (shown by the pins) and the main wetland and floodplain area (in yellow text) 2-9 Figure 2 13: Elevation map and profile of Lower MRB 2-9 Figure 2 14: Estimated population at the ward level for 2012 by resource unit 2-11 Figure 3 1: Summary of the steps involved in the NBI E-Flows Framework (NBI, 2016b) 3-1 17 Figure 3 2: Water governance structure for the reserve in Tanzania 3-5 Figure 3 3: Resource units for the Lower MRB 3-15 Figure 3 4: Methodology flow chart for the water availability assessment in the Lower MRB (SWP, 2019) 3-17 Figure 3 5: Average monthly discharge (left) and runoff regimes (right) at the EFA sites 3-18 Figure 3 6: Predicted river classes of all river segments in Tanzania according to the deductive classification (USAID, 2018). 3-20 Figure 3 7: General ecosystem classification of the resource units in the Lower MRB 3-20 Figure 3 8: Schematic illustration of the main components of a generic flow regime (comprised of individual flow building blocks) considered in the BBM (from CDM Smith, 2016). 3-22 Figure 3 9: Map of the study sites for the social survey, biophysical field campaigns, and RUs 3-30 Figure 4 1: Site photos of Kogatende 4-1 Figure 4 2: Land use within Kogatende sub-catchment (left) and the catchment area (right) 4-2 Figure 4 3: Average monthly, minimum and maximum discharge values for Kogatende 4-2 Figure 4 4: A selection of stage levels for various discharges at Kogatende. Observed discharges indicated by dotted line. 4-3 Figure 4 5: Frequency distribution of depth-velocity classes for the wetted channel for the Mara River at Kogatende 4-3 Figure 4 6: Aerial view of Kogatende site showing the location of various geomorphic features and the location of the cross section. 4-4 Figure 4 7: Cross section of the Kogatende site showing the geomorphic features, sediment deposition and vegetation types. 4-5 Figure 4 8: Cross section of the Mara River at Kogatende showing indicator plant species 4-6 Figure 4 9: Aerial view of Kogatende showing the delineated GHUs for the site (Google Earth) 4-7 Figure 4 10: A photograph depicting habitat characteristics for the Kogatende reach (February, 2019) 4-7 Figure 4 11: Site photos of Tobora 4-13 Figure 4 12: Land use within the Tobora sub-catchment (left) and catchment area (right) 4-16 Figure 4 13: Average monthly, minimum and maximum discharge values for Tobora 4-17 Figure 4 14: Cross section showing observed and modelled flow data for Tobora 4-17 Figure 4 15: Frequency distribution of velocity-depth classes for Tobora 4-18 Figure 4 16: Aerial view of the Tobora River showing the location of the cross section, pools and riffles (Google Earth image 24 September 2010). 4-19 Figure 4 17: A cross sectional view of the Tobora River indicating geomorphological features, sediment composition and vegetation types 4-19 Figure 4 18: Cross section of the Tobora River showing indicator plant species 4-20 Figure 4 19: Aerial view of the Tobora River showing the delineated GHUs for the site (Google Earth) 4-21 Figure 4 20: A photograph depicting habitat characteristics for the Tobora River reach (February 2019) 4-22 Figure 4 21: Site photos of Somoche 4-27 Figure 4 22: Land use within the Somoche sub-catchment (left) and catchment area (right) 4-30 Figure 4 23: Average monthly, minimum and maximum discharge values for Somoche 4-31 Figure 4 24: Cross section showing observed (dotted lines) and modelled (solid lines) water levels for Somoche 4-32 Figure 4 25: Frequency distribution of velocity-depth classes for the wetted Somoche River channel 4-32 Figure 4 26: A satellite image of the Somoche River channel indicating the location of transect across a riffle (Google Earth image 21 July 2017). 4-33 Figure 4 27: A cross section across the Somoche River showing geomorphic features, dominant sediment composition and vegetation types. 4-33 Figure 4 28: Cross section of the Somoche River showing indicator plant species 4-34 Figure 4 29: Aerial view of the Somoche reach showing the delineated GHU for the site (Google Earth) 4-35 Figure 4 30: A photograph depicting habitat characteristics for the Somoche reach (February 2019) 4-36 Figure 4 31: Site photos of Tigithe 4-41 18 Figure 4 32: Land use within the Tigithe sub-catchment (left) and catchment area (right) 4-46 Figure 4 33: Average monthly, minimum and maximum discharge values for the Tigithe EFA site 4-47 Figure 4 34: Cross section for Tigithe River showing observed flow levels (dotted lines) and modelled flow levels (solid lines) 4-47 Figure 4 35: Frequency distribution for velocity-depth classes for the wetted channel of the Tigithe River 4-48 Figure 4 36: A satellite image showing the transect position (white line) and the pool-riffle sequence along the Tigithe River (Google Earth Image date 17 Jan 2010). 4-49 Figure 4 37: Channel cross section for the Tigithe River indicating the geomorphic features (black text), sediment type (brown text) and vegetation type (green text). 4-49 Figure 4 38: A google Earth image showing the back swamp (encircled by the white line) formed by the alluvial ridge/levee (brown line) along Mara River. The Tigithe River drains into the back swamp permanently and the Mara River only during flood flows. Flow direction is from east to west. 4-50 Figure 4 39: Cross section of the Tigithe River showing indicator plant species 4-51 Figure 4 40: Aerial view of the Tigithe River showing the delineated GHUs for the site (Google Earth) 4-52 Figure 4 41: A photograph depicting habitat characteristics for the Tigithe River reach (February 2019) 4-52 Figure 4 42: Site photos of Mara Mines 4-58 Figure 4 43: Land use within the Mara Mines sub-catchment (left) and catchment area (right) 4-61 Figure 4 44: Average monthly, minimum and maximum discharge values for the Mara Mines EFA site 4-61 Figure 4 45: Cross section for the Mara River at Mara Mines for observed (dotted lines) and modelled (solid lines) flow levels 4-62 Figure 4 46: Frequency distribution of depth-velocity classes for the wetted channel of the Mara River floodplain site near Mara Mines 4-62 Figure 4 47: Aerial view of the Mara River at Mara Mines showing the location of the cross section, the meandering planform and sandy bed (Google Earth image date 14 April 2017). 4-63 Figure 4 48: Cross section of the Mara River channel at Mara Mines showing sediment composition, vegetation type and geomorphic features. 4-64 Figure 4 49: Cross section of the Mara River at Mara Mines showing indicator plant species 4-65 Figure 4 50: Aerial view of the Mara River showing the delineated GHUs for the site (Google Earth) for the Mara Mine upstream 4-66 Figure 4 51: Aerial view of the Mara River showing the delineated GHUs for the site (Google Earth) for the Mara Mine downstream 4-67 Figure 4 52: A photograph depicting habitat characteristics for the downstream reach for the Mara Mine site (February 2019) 4-68 Figure 4 53: Site photos of Bisarwi 4-73 Figure 4 54: Land use within the Bisarwi sub-catchment (left) and catchment area (right) 4-76 Figure 4 55: Average monthly, minimum and maximum discharge values for the Bisarwi EFA site 4-76 Figure 4 56: Cross section indicating observed (dotted lines) and modelled (solid lines) flow levels at the Bisarwi EFA site 4-77 Figure 4 57: Frequency distribution of depth-velocity classes for the Mara River at Bisarwi 4-77 Figure 4 58: Aerial view of the Mara River (blue arrows) at Bisarwi indicating the location of the transect and the bifurcation of the Mara River upstream of the avulsion that took place in 1989 (Google Earth image 21 July 2017). The black arrow indicates the new channel and the white arrows indicate flood channels. 4-78 Figure 4 59: Cross section of the Mara River near Bisarwi showing the morphological features, sediment composition and vegetation composition. 4-79 Figure 4 60: Cross section of the Mara River at Bisarwi showing indicator plant species 4-80 Figure 4 61: Aerial view of the Mara River showing the delineated GHUs for the site (Google Earth) 4-81 Figure 4 62: A photograph depicting habitat characteristics for the Mara River reach at Bisarwi (Feb 2019) 4-81
19 Figure 4 63: Site photos of Mara Wetland 4-87 Figure 4 64: Land use within the Mara Wetland sub-catchment (left) and catchment area (right) 4-89 Figure 4 65: Average monthly, minimum and maximum discharge values for the Mara Wetland EFA site 4-90 Figure 4 66: Cross section for the southern extent of the Mara wetland. Dotted lines indicate observed flow levels and solid lines modelled flow levels. The green rectangles indicate the extent of the dense Papyrus plants. 4-90 Figure 4 67: Frequency distribution for velocity-depth classes for the Mara wetland 4-91 Figure 4 68: Plan view of the Mara wetland transect and flow direction (Google Earth image dated 28 December 2018). 4-92 Figure 4 69: Cross section of the left bank side of the Mara wetland indicating the main geomorphological features, sediment composition and vegetation types. Note the wetland extends for another 2 km to the right bank. The dashed line indicate the likely bathometry of the cross section that was not surveyed. 4-92 Figure 4 70: Cross section of the Mara wetland at Ketasakwa showing indicator plant species 4-93 Figure 4 71: Aerial view of the Mara River showing the delineated GHUs for the site (Google Earth) 4-94 Figure 4 72: A photograph depicting habitat characteristics for the Mara River wetland reach 4-95 Figure 5 1: Environmental flow values for Kogatende 5-24 Figure 5 2: Environmental flow values for Tobora 5-28 Figure 5 3: Environmental flow values for Somoche 5-31 Figure 5 4: Environmental flow values for Tigithe 5-34 Figure 5 5: Environmental flow values for Mara Mines 5-37 Figure 5 6: Environmental flow values for Bisarwi 5-40 Figure 5 7: Description of the multi-level monitoring plan (CDM Smith, 2018) 5-43 Figure 5 8: Example of adaptive management cycles for environmental flows from the Rufiji River Basin, Tanzania (CDM Smith, 2018) 5-50
List of Tables Table 2 1: Area of districts inside the Lower MRB 2-2 Table 2 2: Hydrological units in the Lower MRB 2-4 Table 2 3: District population growth rates 2-10 Table 2 4: Population data by district 2-10 Table 2 5: Population data by HU 2-10 Table 3 1: List of tasks recommended and undertaken for each step of the NBI E-flows Framework 3-2 Table 3 2: Domestic water requirements for water resources planning in Tanzania 3-7 Table 3 3: Catchment characteristics of the EFA sub-catchments 3-17 Table 3 4: Summary of frequency analysis for maximum daily flow values 3-18 Table 3 5: Ecosystem classification details for each resource unit and associated EFA study site 3-21 Table 3 6: The HCR classifications for respective flow velocity and recorded depths 3-25 Table 3 7: Selected parameters for Tanzania Drinking Water Standards 3-27 Table 3 8: Selected parameters for Tanzania Municipal and Industrial Wastewaters 3-27 Table 3 9: Example physico-chemical benchmarks for freshwater ecosystems (UNEP/UNU-EHS, 2016) 3-28 Table 3 10: List of the corresponding EFA study site names and RUs 3-30 Table 3 11: Location details for the biophysical study sites 3-31 Table 3 12: Biophysical study site description (transect, location and flow) 3-32 Table 4 1: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding details 4-6 Table 4 2: Macroinvertebrate community metrics for Kogatende 4-8 Table 4 3: Results of the SASS5 and TARISS on processed samples at Kogatende 4-8 Table 4 4: Results of water quality analysis at Kogatende 4-9 Table 4 5: Site metrics for Kogatende 4-9 Table 4 6: Indicators and management objectives for Kogatende 4-10
20 Table 4 7: Required flow conditions for Kogatende 4-11 Table 4 8: Confidence level for flow recommendations for Kogatende 4-12 Table 4 9: Wetland resources and their relative importance to the livelihood of communities in Tobora 4-14 Table 4 10: Economic activities in Tobora RU 4-15 Table 4 11: Trend of resources and conditions of the wetland in Tobora 4-15 Table 4 12: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding d e t a i l s . 4 - 2 1 Table 4 13: Macroinvertebrate community metrics for Tobora site 4-23 Table 4 14: Results of the SASS5 and TARISS on processed samples at Tobora 4-23 Table 4 15: Results of water quality analysis at Tobora 4-24 Table 4 16: Site metrics for Tobora 4-24 Table 4 17: Indicators and management objectives for Tobora 4-24 Table 4 18: Required flow conditions for Tobora 4-25 Table 4 19: Confidence level for flow recommendations for Tobora 4-26 Table 4 20: Wetland resources and their relative importance to the livelihood of communities in Somoche RU 4-28 Table 4 21: Economic activities in Somoche RU 4-29 Table 4 22: Trend of resource utilization and condition of the wetland in Somoche RU 4-29 Table 4 23: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding d e t a i l s . 4 - 3 5 Table 4 24: Macroinvertebrate community metrics for Somoche 4-36 Table 4 25: Results of the SASS5 and TARISS on processed samples at Somoche 4-37 Table 4 26: Results of water quality analysis at Somoche 4-37 Table 4 27: Site metrics for Somoche 4-37 Table 4 28: Indicators and management objectives for Somoche 4-38 Table 4 29: Required flow conditions for Somoche 4-39 Table 4 30: Confidence level for flow recommendations for Somoche 4-40 Table 4 31: Wetland resources and their relative importance to the livelihood of communities in Lower Tigithe 4-42 Table 4 32: Economic activities in Lower Tigithe RU 4-43 Table 4 33: Trend of resources and condition of the wetland in Lower Tigithe 4-43 Table 4 34: Wetland resources and their relative importance to the livelihood of communities in Upper Tigithe 4-44 Table 4 35: Economic activities in Kitawasi village in Upper Tigithe 4-45 Table 4 36: Trend of resources utilization and condition of the wetland in Upper Tigithe 4-45 Table 4 37: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding details. 4-51 Table 4 38: Macroinvertebrate community metrics for Tigithe 4-53 Table 4 39: Results of the SASS5 and TARISS on processed samples at Tigithe 4-53 Table 4 40: Results of water quality analysis at Tigithe 4-54 Table 4 41: Site metrics at Tigithe 4-54 Table 4 42: Indicators and management objectives for Tigithe 4-54 Table 4 43: Required flow conditions for Tigithe 4-56 Table 4 44: Confidence level for flow recommendations for Tigithe 4-57 Table 4 45: Wetland resources and their relative importance to livelihoods in Mara Mines RU 4-59 Table 4 46: Economic activities in Mara Mines RU 4-59 Table 4 47: Trend of resources and condition of the wetland in Mara Mines RU 4-60 Table 4 48: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding details for the upstream reach. 4-66 Table 4 49: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding details for the downstream reach. 4-67 Table 4 50: Macroinvertebrate community metrics for Mara Mines 4-68 Table 4 51: Results of the SASS5 and TARISS on processed samples at Mara Mines 4-69 Table 4 52: Results of the water quality analysis for Mara Mines 4-69
21 Table 4 53: Site metrics for Mara Mines 4-69 Table 4 54: Indicators and management objectives for Mara Mines 4-70 Table 4 55: Required flow conditions for Mara Mines 4-71 Table 4 56: Confidence level for flow recommendations for Mara Mines 4-72 Table 4 57: Wetland resources and their relative importance in North Mara RU 4-74 Table 4 58: Economic activities in North Mara RU 4-74 Table 4 59: Trend of resources and condition of the wetland in North Mara RU 4-75 Table 4 60: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding details. 4-80 Table 4 61: Macroinvertebrate community metrics for Bisarwi site 4-82 Table 4 62: Results of the SASS5 and TARISS on processed samples at Bisarwi site 4-82 Table 4 63: Results of the water quality analysis at Bisarwi 4-83 Table 4 64: Site metrics for Bisarwi 4-83 Table 4 65: Indicators and management objectives for Bisarwi 4-83 Table 4 66: Required flow conditions for Bisarwi 4-84 Table 4 67: Confidence level for flow recommendations for Bisarwi 4-85 Table 4 68: Wetland resources and their relative importance to the livelihood of communities in South Mara RU 4-87 Table 4 69: Economic activities in South Mara RU 4-88 Table 4 70: Trend of resources and condition of the wetland in South Mara RU 4-88 Table 4 71: A summary of the HCR for the reach with associated velocity depth class ratings and corresponding details. 4-94 Table 4 72: Macroinvertebrate community metrics for the Mara Wetland 4-95 Table 4 73: Results of the SASS5 and TARISS on processed samples for the Mara Wetland 4-96 Table 4 74: Results of the water quality analysis at the Mara Wetland 4-96 Table 4 75: Site metrics for the Mara Wetland 4-96 Table 4 76: Indicators and management objectives for the Mara Wetland 4-97 Table 4 77: Required flow conditions for the Mara Wetland 4-98 Table 4 78: Confidence level for flow recommendations for Mara Wetland 4-99 Table 5 1: Final RQO statements for Kogatende 5-2 Table 5 2: RQO targets and indicators for Kogatende 5-2 Table 5 3: Final RQO statements for Tobora 5-4 Table 5 4: RQO targets and indicators for Tobora 5-4 Table 5 5: Final RQO statements for Somoche 5-6 Table 5 6: RQO targets and indicators for Somoche 5-6 Table 5 7: Final RQO statements for Tigithe 5-8 Table 5 8: RQO targets and indicators for Tigithe 5-9 Table 5 9: Final RQO statements for Mara Mines 5-10 Table 5 10: RQO targets and indicators for Mara Mines 5-11 Table 5 11: Final RQO statements for Bisarwi 5-13 Table 5 12: RQO targets and indicators for Bisarwi 5-13 Table 5 13: Final RQO statements for Mara Wetland 5-14 Table 5 14: Indicators and management objectives for the Mara Wetland 5-15 Table 5 15: Basic human needs estimates for 2018 by resource unit 5-16 Table 5 16: Future basic human needs estimates by resource unit 5-17 Table 5 17: Environmental flow values for a maintenance year 5-19 Table 5 18: Environmental flow values for drought year 5-20 Table 5 19: Environmental flow values for Kogatende 5-23 Table 5 20: Freshet and flood events for environmental flows for Kogatende 5-24 Table 5 21: Environmental flow values for Tobora 5-27 Table 5 22: Freshet and flood events for environmental flows for Tobora 5-28 Table 5 23: Environmental flow values for Somoche 5-30 Table 5 24: Freshet and flood events for environmental flows for Somoche 5-31 Table 5 25: Environmental flow values for Tigithe 5-33 Table 5 26: Freshet and flood events for environmental flows for Tigithe 5-34
22 Table 5 27: Environmental flow values for Mara Mines 5-36 Table 5 28: Freshet and flood events for environmental flows for Mara Mines 5-37 Table 5 29: Environmental flow values for Bisarwi 5-39 Table 5 30: Freshet and flood events for environmental flows for Bisarwi 5-40 Table 5 31: Reserve monitoring recommendations 5-46
List of Acronyms and Abbreviations
Acronym Definition ASPT average score per taxon BBM Building Block Methodology DO dissolved oxygen EC electrical conductivity EFA environmental flow assessment EIS Ecological Importance and Sensitivity EMC Ecological Management Class FD fast deep (habitat) FS fast shallow (habitat) HCR Habitat Cover Rating HU hydrological unit IHE Delft IHE Delft Institute for Water Education GHU Geomorphic Habitat Units GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit GLOWS-FIU Global Water for Sustainability program, Florida International University km kilometer LVBWB Lake Victoria Basin Water Board m/s meters per second m3/s cubic meters per second m meter MaMaSe Mau Mara Serengeti Sustainable Water Mm3 million cubic meters mm millimeter MoU Memorandum of Understanding MoW Ministry of Water MRB Mara River Basin NBI Nile Basin Initiative NELSAP-CU Nile Equatorial Lakes Subsidiary Action Program Coordination Unit PREPARED Planning for Resilience in East Africa through Policy, Adaptation, Research and Economic Development PES Present Ecological State RQO Resource Quality Objective RU resource unit SASS/SASS5 South African Scoring System SENAPA Serengeti National Park SIS Social Importance and Sensitivity SD slow deep (habitat) SS slow shallow (habitat) TARISS Tanzania River Scoring System ToC Trajectory of Change USAID United States Agency for International Development WAP water allocation plan WUA water users association
23 24 1. INTRODUCTION
WWF World Wide Fund for Nature MRB. These are related components of sustainable yr year water resource management featured in management frameworks for the Nile Basin as well as Tanzanian 1.1 Project Scope and Objectives laws and regulations. RQOs and the reserve setting This report presents the results of the determination of is a requirement as stipulated in Part VI (Protection environmental flow requirements in the Lower Mara of Water Resources) of the 2009 National Water River Basin. The Lower Mara Environmental Flow Resources Management Act of Tanzania. RQOs Assessment (EFA) was carried out by The Nile Basin are intended to protect water and related aquatic Initiative, Nile Equatorial Lakes Subsidiary Program biological resources at levels needed to meet the (NBI/NELSAP) in collaboration with Deutsche needs of resource users and maintain ecosystems in Gesellschaft für Internationale Zusammenarbeit a desired environmental management class. (GIZ) through technical support provided by IHE Delft Institute for Water Education (IHE Delft) under The detailed aspects of RQOs are not specified by the the project Biodiversity Conservation and Sustainable Tanzanian MoW but are described in the technical Utilization of Ecosystem Services of Wetlands of manual accompanying the NBI E-Flows Strategy Transboundary Relevance in the Nile Basin, which (NBI, 2016a) which have been adopted by the United seeks to build consensus and enhance cooperation Republic of Tanzania through its membership in in water resources management and development the NBI. The NBI E-Flows Framework adopts the between the Nile Basin’s riparian countries. description of RQOs specified in the 2004 South African National Water Resource Management The objective of the project is to determine the flow Strategy , which states that “resource quality includes requirements needed to meet the reserve in the water quantity and water quality, the character and Tanzanian part of the Mara River Basin (referred condition of in-stream and riparian habitats, and to in this document as the Lower Mara River Basin, the characteristics, condition and distribution of the or Lower MRB) using the Nile E-flows Framework aquatic biota. developed by NBI. The study builds upon a similar EFA study conducted by NBI in cooperation with Environmental flows and the reserve are closely the Mau Mara Serengeti Sustainable Water Initiative related instruments to meet RQOs. The Nile E-flows (MaMaSe) project in the Kenyan part of the basin Strategy defines environmental flows as describing (more details in Section 3.1.3). Mara River is a “the quantity, timing and quality of water flows transboundary river originating from the partially required to sustain freshwater and estuarine deforested Mau Escarpment of Kenya, flowing through ecosystems and the human livelihoods and well- the savanna landscapes of Maasai Mara National being that depend upon these ecosystems”. This Reserve and Serengeti National Park (SENAPA), and definition is consistent with the internationally emptying into Lake Victoria via an extensive wetland recognized definition included in the 2007 Brisbane of high conservation value. Despite its modest area Declaration . The term environmental flows is not (13,750 km2), the basin contains a wide variety of used in Tanzanian water laws or regulations. Instead ecosystem types characteristic of some of the most the 2009 National Water Resources Management Act valued natural features of the Nile Basin. of Tanzania refers to the “reserve”, which is defined as the quantity and quality of water required for - The Mara is also home to approximately 1 million inhabitants dependent on the river’s water resources (a) satisfying basic human needs by securing a basic for a wide variety of livelihoods. These characteristics, water supply for people who are now or who shall in in additional to the relative abundance of past studies the reasonably for near future, be and available data, make the Mara a valuable model (i) relying upon for learning and demonstration in the wider Lake (ii) taking water from; or Victoria and Nile Basin context. More details on the (iii) being supplied from the relevant water resources; biophysical and social characteristics of the basin and are presented in Chapter 2. The effort was planned and carried out in close cooperation with and direct (b) protecting to protect [sic] aquatic ecosystem in involvement of the Lake Victoria Basin Water Board order to secure ecologically sustainable development (LVBWB) and the Ministry of Water (MoW) of and use of the relevant water resources. Tanzania, which is the authority legally responsible for setting and protecting environmental flows in The reserve thus consists of two parts. Part 1 is water resource management. Officers of LVBWB focused on meeting basic human needs, which can were engaged in every aspect of the project and MoW be considered a component of the domestic water task force members were engaged during all major demand, and part 2 is focused on protecting aquatic project activities, including field work, stakeholder ecosystems. In this study we distinguish between the workshops, and technical meetings. basic human need component of the reserve and the ecological component of the reserve, referred to as 1.2 Resource Quality Objectives, Environmental Flows and the Reserve This study focuses on setting RQOs and quantification of environmental flows and the reserve in the Lower 25 environmental flows. More details on the Tanzanian document will act as a direct input into both the policy context for this study are presented in Section Tanzanian and transboundary WAP efforts. 3.1.1. 1.3 Cooperation with National and 1.4 Content of this Report Transboundary Water Allocation This report consists of eight chapters. Following Planning Activities this introduction chapter, Chapter 2 presents This report is a complete and independent report an overview of the features of the Lower MRB, documenting the details of the reserve determination including its administrative boundaries, process for the Lower MRB. However, this project hydrology, climate, land use, physiography and was developed and carried out in close cooperation geomorphology, basin communities and social with parallel activities of the LVBWB to develop a importance. Chapter 3 presents the methodology WAP for the Lower MRB. A central component of used in the assessment, addressing each of the the WAP is the water balance of the basin, which is seven phases of the NBI E-flows Framework. defined as the difference between the total available Although Tanzania has developed draft water in the basin and the sum of the reserve and guidelines on the methods to be used for assessing demands of other water users. Under a parallel effort environmental water requirements, these have by the Sustainable Water Partnership financed by the not been approved. In consultation with the United States Agency for International Development Tanzanian water authorities we have chosen to (USAID), IHE Delft is supporting the LVBWB to apply the framework developed by the NBI, as quantify the total available water and demands of this also meets the highest level of methodologies other users in the Lower MRB. World Wide Fund proposed in the draft Tanzanian guidelines. for Nature (WWF) Tanzania is also supporting the stakeholder engagement aspects of this effort. Chapter 4 describes key features of the seven sites chosen for detailed assessment in this The results of this EFA will therefore be carried study. These include two mainstem sites, three forward to become an input to the Tanzania Mara tributary sites, and two sites in Mara Wetland. WAP. (More details on the WAP process, including Features described include surveyed social transboundary aspects, are presented in Section data and observed characteristics of hydrology, 3.1.4). Upon its completion, efforts will be made to hydraulics, geomorphology, riparian vegetation, harmonize the Tanzania Mara WAP with a similar macroinvertebrates, fish, and water quality plan that has been developed and is undergoing at each site. Chapter 5 presents the reserve modification for the Kenya part of the basin. The assessment results, organized according to the ambition is to develop a single, transboundary WAP seven phases of the Nile E-flows Framework that can be agreed by Tanzania and Kenya within the and includes the final results of the RQO framework of the Memorandum of Understanding process, quantification of basic human needs (MoU) for Joint Water Resources Management of the requirements, and the environmental flow values Transboundary Mara River Basin signed between and motivations for each EFA study site. Chapter Kenya and Tanzania in September 2015. As such, this 6 includes discussions on specific topics including uncertainties encountered while carrying out the
1 http://www.dwa.gov.za/Documents/Policies/NWRS/Default.htm 2 The 2007 Brisbane Declaration is accessible via http://riverfoundation.org.au/wp-content/uploads/2017/02/THE-BRISBANE-DECLARATION.pdf 26 27 1. OVERVIEW OF THE LOWER MARA RIVER BASIN
NBI E-Flows Framework. Chapters 7 and 8 present round source of water in the Mara River. It passes references and annexed material, respectively. through villages, wildlife conservancies, and Maasai The MRB basin is a transboundary river basin located Mara National Reserve in Kenya, and then through in Kenya and Tanzania. The entire MRB covers SENAPA, more villages, gold mining operations, and about 13,750 km2 and begins in the Mau Forest in the Mara Wetland in Tanzania before flowing into Kenya which feed two perennial rivers, the Amala Lake Victoria. The Lower MRB is the part of the MRB and Nyangores, which provide for the main year-
FIGURE 2 1: MAP OF THE MARA RIVER BASIN
28 located downstream of the Kenya-Tanzania border and covers an area of 5,047 km2. water use fees, and engagement of communities on 2.1 Administration water resources management. The main office for the The Lower MRB is included within the Lake Victoria LVBWB is located in Mwanza, with two sub-offices Basin, which is one of nine major river basins in in Bukoba and Musoma. The Musoma sub-office is Tanzania and is managed by the LVBWB. The where the primary location for management decisions LVBWB was formed in 2000 with the role of water and actions related to the Lower MRB. Within the allocation and pollution control, issuing of water Lower MRB, a Mara Subcatchment Committee as use and discharge permits, billing and collection of
FIGURE 2 2: MAJOR RIVER BASINS LOCATED IN TANZANIA well as six water user associations (WUAs) have been Districts (4 percent). Each district has its own district established. governments, including district water engineers and The Lower MRB is also located within the Mara environmental officers, while the Mara Region has Region in Tanzania and includes four districts (Figure its own government officials located in Musoma, 2 3, Table 2 1). Serengeti District contains the largest portion of the Lower MRB (65 percent), followed by Butiama (16 percent), Tarime (15 percent), and Rorya
29 FIGURE 2 3: DISTRICTS LOCATED IN THE LOWER MRB
30 including the regional administrative secretary and zonal irrigation office. passes the gauging station at the Kirumi Bridge, 2.2 Hydrology and enters Lake Victoria at the Mara Bay close to The Mara River enters Tanzania at Purungat on the Musoma. border of Maasai Mara National Reserve in Kenya As part of previous and on-going efforts related to and SENAPA in Tanzania. It is joined by a seasonal environmental flows and water allocation planning, tributary called the Sand River only a couple of the entire Mara River Basin has been divided into hundred metres downstream. The river flows through hydrological units (HUs). These HUs combine SENAPA and passes the gauging station Kogatende. drainage areas that share similar characteristics in Shortly after leaving SENAPA, it is joined by the topography and rainfall patterns. The Lower MRB Tobora River, a tributary flowing into the mainstem contains three HUs: Serengeti, Somoche and Mara from the south. Beyond the bridge at the Tarime- (Figure 2 4). Only a very small portion of a fourth Mugumu road, with the gauging station Nyasurura, HU; Sand, lies within Tanzania and was not included the Somoche River (largest southern tributary) joins in this effort. Table 2 2 provides a summary of the the Mara River. At this point the river course turns characteristics of the HUs including their long-term north, passing the gauging station at Mara Mines, average water balance values. Annual precipitation and then turns east again before passing the North in the Lower MRB ranges between 926 and 1,009 Mara Gold Mine. mm, generating an annual runoff of 45 to 54 mm. The resulting annual actual evaporation from the three There the Mara flows into Mara Wetland. The Tigithe sub-catchments varies between 881 and 956 mm. River, an important northern tributary, enters the Mara Wetland east of Bisarwi. Flow in the 250 km2 Based on the lower precipitation amounts within Mara Wetland is dispersed in multiple channels that the Lower MRB compared to the values in the become obscured by vegetation and indistinct in the upper part of the catchment, the runoff contribution wetland core. Near the western margin, the river
FIGURE 2 4: HUS CONTAINED IN THE LOWER MRB
31 accumulates to only 24 percent of the total runoff, despite the relatively large areal extent of the Lower the highest precipitation, whereas the southern MRB, which is 38 percent of the total area. and western parts receive considerably lower 2.3 Climate precipitation (Figure 2 5 and Figure 2 6). Monthly The annual average precipitation of stations in and precipitation data clearly indicate a bimodal regime around the Lower MRB varies from 680 mm at Musoma to 1,336 mm at Kichwa Tembo. The northern and north-eastern parts of the catchment receive
FIGURE 2 5: SPATIAL DISTRIBUTION OF PRECIPITATION STATIONS IN THE LOWER MRB AND THIESSEN POLYGON AREAS
32 FIGURE 2 6: AVERAGE ANNUAL PRECIPITATION AT STATIONS RELEVANT TO THE LOWER MRB
FIGURE 2 7: AVERAGE MONTHLY DISCHARGE (LEFT) AND RUNOFF REGIMES (RIGHT) AT HU1 OUTLETS
33 FIGURE 2 8: AVERAGE MONTHLY PRECIPITATION AT STATIONS RELEVANT FOR THE LOWER MRB with two rainfall seasons. The long rains occur from The highest monthly flows occur in April and May; March to May and the shorts rains from November lowest flows are in August for the Somoche HU2, and to December while dry seasons are experienced from in October for the Mara and Serengeti HUs (Figure June to October and January to February. 2 7). The intra-annual variability of flows in the Lower Somoche HU2 is only influenced by local precipitation MRB reflects the rainfall pattern in the catchment as inputs, whereas Serengeti HU3 and Mara HU2, on well as the dependency of flows from the upstream the mainstem of the Mara, also receive flows from catchment in the case of the mainstem Mara River. the upper parts of the catchment. This difference can be seen in the right graph of Figure 2 8. This graph
34 shows the runoff regimes using the Pardé coefficients. Pardé coefficients are calculated by dividing the mean
FIGURE 2 9: LAND USE MAP OF THE MARA RIVER BASIN (MAMASE, 2019) monthly value by the annual average. savannah and grassland vegetation. Some large-scale irrigation farms are located around this area and 2.4 Land Use crops grown are mostly cereals, French beans and Land use in the basin varies significantly from the avocados for export. This savanna area stretches into headwaters in Kenya to the lower basin in Tanzania the protected areas of the well-known Maasai Mara (Figure 2 9). The most upstream part of the basin, National Reserve and SENAPA. These protected where the river originates, is mostly made up of areas are home to a variety of wild animals and attract forest plantations. Further down, a buffer zone tourists from all over the world. comprising of tea estates separates the forest from the agricultural lands and was meant to prevent A mixture of small and large-scale gold mines, encroachment by the locals into the forested areas. agricultural areas and livestock dominate the areas Small scale, rainfed, subsistence farming is mostly west of the SENAPA. Before the Mara River flows carried out in areas near the tea plantations. A variety into the lake, it enters the floodplains of the Mara of crops are grown including maize, potatoes, beans, Wetland (covering 17 percent of the Mara HU1, vegetables etc. Livestock farming is also practiced in Figure 2 10) which plays a significant role in terms these areas although at a smaller scale. Some small of trapping sediments flowing from the upstream to medium sized urban centers are found within the areas, purifying water, providing habitat for aquatic basin, namely Olenguruone, Bomet, Mulot, Longisa, organisms and birds, and recharging ground water and Talek in Kenya and Mugumu in Tanzania. among other services. The wetland provides food (fish) for local populations, papyrus for making mats The middle part of the basin is mostly dominated by and other artefacts and water for irrigating small
35 FIGURE 2 10: LAND USE CLASSES WITHIN THE LOWER MRB CATCHMENTS HU1 (LEFT), HU2 (CENTRE), AND HU3 (RIGHT) farms nearby.
the Eastern Rift Valley. This might have rejuvenated, 2.5 Physiography and straightened and incised the Mara River since the Geomorphology formation of Eastern Rift Valley. This rejuvenation is The Lower MRB is situated along a half-graben that visible along the longitudinal profile where it forms a has been influenced by tectonic activity related to convex (see 220-280 km along the profile in Figure
FIGURE 2 11: A LONGITUDINAL PROFILE OF THE MARA RIVER SHOWING THE EFA SITES OF 2016 AS ORANGE DOTS AND THE CURRENT BIOPHYSICAL STUDY SITES AS BLACK DOTS
2 11). The base-level control at the lower-end of the deposition and dense vegetation growth. This led to system is provided by Lake Victoria. the formation of the Mara floodplain and the Mara Wetland (Figure 2 12). This base-level control created a low gradient, The elevation of the Lower Mara River varies from low energy environment that promoted sediment 1,400 m at the Kenya-Tanzania border to 1,133 m at the mouth where the river flows into the Lake Victoria
36 FIGURE 2 12: LOCATION OF THE SAMPLING SITES (SHOWN BY THE PINS) AND THE MAIN WETLAND AND FLOODPLAIN AREA (IN YELLOW TEXT)
FIGURE 2 13: ELEVATION MAP AND PROFILE OF LOWER MRB
37 (Figure 2 13). The average elevations are 1,517 m in Serengeti HU3, 1,419 m in Somoche HU2, and 1,277 m in Mara HU1. and fishing, and this varies from village to village. 2.6 Communities and Social The estimated population for the Lower Mara was Importance 335,000 in 2012. The population growth for each The dominant tribes in the Lower MRB are the district varies but is between 2 and 3.5 percent (Table Kurias, mainly found in the upper part of the basin, 2 3). Based on the 2012 census data and growth followed by the Luos who mostly reside along the rates, the 2018 estimated population is 396,000. The wetland in the lower part of the basin. Migration population estimates for the districts and HUs are from outside is uncommon although some locals provided in Table 2 4 and Table 2 5, respectively. It move from rural areas to other rural areas. Economic should be noted that the population estimate is for activities carried out in this area include mining, the proportion of the districts located within the sand harvesting, livestock keeping, agriculture
38 FIGURE 2 14: ESTIMATED POPULATION AT THE WARD LEVEL FOR 2012 BY RESOURCE UNIT
Lower Mara and not the entire district. Figure 2 14 However, the importance of fish to the communities’ shows the estimated 2012 population by ward and livelihood is relatively minimal (less than one percent). separated into the different RUs. Fish is generally of low importance compared to other The wetland, including areas along the river banks resources, except in the lower parts of Mara Wetland is an important area for farming activities, especially where it is of higher importance. This is because in the lower Tigithe. Flooding is the main challenge there is relatively more fishing in the lower part of to farming in these areas, critically affecting the the basin. Livestock keeping constitutes one of the lower part of the basin. Unlike other areas, limited most important economic activities in the basin. crop farming is practiced in some parts of the upper Wetland pastures are key resources of importance for Tigithe. This is as a result of widespread planting of livestock. Roofing grass and wood fuel are harvested exotic trees (eucalyptus) which has largely substituted by most communities. Other resources such as other economic activities. natural vegetables, building sand, natural fruits, building stones, weaving grass and medicinal plants With the exception of the Upper Tigithe, fish is commonly harvested in the rivers and natural ponds.
39 40 3. RESERVE ASSESSMENT METHODOLOGY are also collected. Most of these resources can also was developed by NBI to ensure a standard process be obtained in areas away from wetlands and rivers. is followed for the increasing number of EFAs being Building poles are another important resource. conducted in the Nile Basin. There are seven main The determination of the reserve and the related RQOs steps in this framework, which are summarized in was completed using the NBI Environmental Flows Figure 3 1 and Table 3 1 and described in detail in Framework (NBI, 2016a, 2016b). This framework this chapter. Discussions on specific topics, including
FIGURE 3 1: SUMMARY OF THE STEPS INVOLVED IN THE NBI E-FLOWS FRAMEWORK (NBI, 2016B)
41 TABLE 3 1: LIST OF TASKS RECOMMENDED AND UNDERTAKEN FOR EACH STEP OF THE NBI E-FLOWS FRAMEWORK
TABLE 3 1: LIST OF TASKS RECOMMENDED AND UNDERTAKEN FOR EACH STEP OF THE NBI E-FLOWS FRAMEWORK
42 TABLE 3 1: LIST OF TASKS RECOMMENDED AND UNDERTAKEN FOR EACH STEP OF THE NBI E-FLOWS FRAMEWORK
uncertainties encountered during the process, have been included in Section 6. Finally, important local and regional partners were 3.1 Basin Scale Situation identified and included in all relevant project activities Assessment and Alignment Process to promote cooperation and capacity building. Before any physical studies are undertaken for 3.1.1 Tanzania Policy Review determining environmental flows, it is important to There are a variety of pieces of legislation and have a good understanding of the current legislative guidance documents related to the environmental and management mandates. To do this, a review flows and the reserve in Tanzania. The following of the policies in Tanzania was conducted to better policy review divided these documents into three understand the national and regional requirements groups: regulations related to water in Tanzania, and how environmental flows could be properly regulations related to the environment in Tanzania, implemented in Tanzania. and international agreements. For each section, the documents have been arranged in chronological order A review of existing information related to from oldest to newest. For national regulations, they environmental flows and water resources have also been organized in hierarchical structure management was also completed so previously from most broad to most specific (i.e., policy, strategy, collected information could be utilized and project programme, act, and manual/guideline). This policy activities could be aligned with on-going projects. review outlines the water governance structure with
43 FIGURE 3 2: WATER GOVERNANCE STRUCTURE FOR THE RESERVE IN TANZANIA
Tanzania related to the reserve in the Lower MRB as protecting and conserving river and estuary well as important international agreements Tanzania ecosystems as well as the plants and animals that is a signatory to in the region (Figure 3 2). depend on them. 3.1.1.1 Regulations Related to Water National Water Sector Development Strategy, 2006 to 2015 National Water Policy, 2002 The National Water Sector Development Strategy The National Water Policy (URT, 2002) sets out long- (URT, 2008) is the blueprint for implementing the term objectives for water resources management National Water Policy, laying the framework for water across the country, with a specific focus on rural resources assessment, planning, and development at water supply, urban water supply and sanitation, and the local, national, and international scale. For each how the development of water resources intersects water resources management component, it defines a with the economic development of water-dependent problem statement, policy direction, goals, strategies, sectors. With regard to water and the environment, and activities to guide future implementation. the National Water Policy contains the following objective: Environmental Protection and Conservation Goal: Increased environmental protection “To have in place water management system which and conservation measures contribute to the protects the environment, ecological system and sustainability of all aspects of water development, biodiversity...Water for the environment, in terms of management and use. quantity and quality, and levels, and for both surface and groundwater resource shall be determined on Activity: Determine environmental flow requirements the best scientific information available considering for ecosystems for all key rivers. both the temporal and spatial water requirements to maintain the health and viability of riverine and Water Utilization and Allocation estuary ecosystems, and associated flora and fauna.“ Goal: Implementation of a responsive, effective and sustainable water resources utilisation and This lays the foundation for the importance of allocation system based on social and economic priorities whilst maintaining minimum reserves for 44 the protection of eco-systems. Activity: Develop water allocation criteria, procedures and guidelines for water basins. water boards (including the LVBWB), subcatchment Water Sector Development Programme, 2006 committees, and WUAs. It also provides specific – 2025 details on how water is to be managed, including The Water Sector Development Programme (URT, permitting, fees, protected areas, and risk 2006) is a direct output of the National Water management. The Water Resources Management Act Sector Development Strategy and is critical for the includes the formal definition of the reserve: implementation of the National Water Policy. It combines the strategies for the three sub-sectors The “reserve“ means the quantity and quality of water outlined in the National Water Policy: Water required for - Resources Management, Rural Water Supply and (a) satisfying basic human needs by securing a basic Sanitation, and Urban Water Supply and Sewerage. water supply for people who are now or who shall in the reasonably for near future, be Specifically, the Water Resources Management sub- (i) relying upon sector is the most extensive of the three and supports (ii) taking water from; or the strengthening of basin water offices (also known (iii) being supplied from the relevant water resources; as basin water boards) in their efforts for water and resources monitoring, assessment, and enforcement. Important activities related to environmental flows (b) protecting to protect [sic] aquatic ecosystem in includes the protection of important water sources, order to secure ecologically sustainable development including to “empower the Minister to establish and and use of the relevant water resources. set aside a “reserve” before water allocation decisions are made”. It also notes that specific operational Design Manual for Water Supply and Waste manuals and guidelines should be created to provide Water Disposal, 2009 advice for topics specific to each sub-sector. The MoW of the United Republic of Tanzania has published and regularly updates a design manual Water Resources Management Act, 2009 that provides standard values for water resources The Water Resources Management Act (URT, 2009) planning and development that can be used across establishes the hierarchical government structure for the country (MoWI, 2009). In this design manual are water management as prescribed in the documents numerical values for domestic water requirements above, including the Minister of Water, the Director for different types of areas (rural vs. urban), income of Water Resources, the National Water Board, basin levels, and the type of payment or tariff structure
TABLE 3 2: DOMESTIC WATER REQUIREMENTS FOR WATER RESOURCES PLANNING IN TANZANIA
45 available (Volume 4, Design of Piped Water Systems, other water needs, to protect water sources and Page 4-13). The minimum amount of domestic water to prevent environmental pollution. In order to to be supplied is 25 liters/person/day. achieve this, the following policy objectives shall be While Tanzania does not define the amount required pursued...planning and implementation of water for basic human needs, this value of 25 liters/person/ resources and other development programmes in an day will be used in this effort. This also aligns with the integrated manner and in ways that protect water recommendations provided in the draft Guidelines catchment areas and their vegetation cover.“ for Water Allocation Planning (see below). Environmental Management Act, 2004 Environmental Water Requirements “Basin Water Boards in prioritizing different uses Assessment Guidelines for Tanzania (Final of water shall ensure that adequate water is made Draft, 2016) available for environmental purposes.“ Guidelines for determining environmental water requirements (also known as environmental 3.1.1.3 Regional and International flows) were developed to help basin water boards Agreements determine the aquatic ecosystem protection requirements of the reserve as defined in the Water East African Community and the Lake Resources Management Act of 2009 (URT, 2016). Victoria Basin Commission It recommends methodologies for different types The East African Community is a regional of water bodies, including rivers, lakes, estuaries, intergovernmental organisation made up of six and wetlands. It suggests methodologies for rapid countries in the east African region: the Republics of assessments (level 1) as well as more detailed holistic Burundi, Kenya, Rwanda, South Sudan, the United assessments (level 2). Level 2 assessments should be Republic of Tanzania, and the Republic of Uganda. completed “for specific rivers or river reaches where It was established in 1999 with the mission “to such existing environmental problems are caused by widen and deepen economic, political, social and anthropogenic activities and a compromise is needed cultural integration in order to improve the quality between environmental health of rivers and human of life of the people of East Africa through increased development”. It outlines potential methodologies competitiveness, value added production, trade and that could be used (including the BBM) but suggests investments”. This is carried out across a variety that any methodology could be used as long as of sectors including the environment and natural it is capable of providing environmental water resources. In particular, it is important the East requirement values for different management classes. African Community member states cooperate in the These guidelines are still under review by the MoW sustainable management of “biologically significant and have yet to be finalized and approved. transboundary freshwater ecosystems”, which includes ecosystems like the Mara River Basin. Guidelines for Water Allocation Planning (Draft, 2018) A specialized institution within the East African These guidelines were drafted in May 2018 and outline Community is the LVBC, which has the special specific methodologies and considerations when basin mission of coordinating sustainable development and water boards develop their water allocation plan, management of resources within the countries that as mandated in the Water Resources Management are included in the Lake Victoria Basin. They have the Act of 2009 (URT, 2018a). It provides guidance on role of providing neutral oversight and coordination the quantification of the reserve, including how to between countries when working on transboundary calculate water requirements for basic human needs issues related to water resources management. and for the environmental component of the reserve, which follows the recommendations provided in the Nile Basin Initiative and Nile Equatorial Environmental Water Requirements Assessment Lakes Subsidiary Action Program Guidelines for Tanzania. The WAP guidelines are still Tanzania joined the NBI in 1999 as one of the original under development and have yet to be finalized and nine countries to create the partnership. Through approved by the MoW. this partnership, it agrees to follow transboundary water management strategies developed by NBI, 3.1.1.2 Regulations Related to the which include the Wetland Management Strategy Environment (NBI, 2013) and the Strategy for Management of While less directly related to the reserve, the national Environmental Flows in the Nile Basin (NBI, 2016b). legislative documents for environmental management The Strategy for Management of Environmental also provide legal support for the implementation, Flows in the Nile Basin is being applied to the MRB monitoring, and adaptive management of the due to its transboundary status with Kenya. reserve. The two main documents in this sector are the National Environmental Policy (URT, 1997) and NELSAP is one of two investment programs under NBI the Environmental Management Act (URT, 2004). with the mission of “to contribute to the eradication of poverty, promotion of economic growth, and National Environmental Policy, 1997 reversal of environmental degradation” in the Nile “The environmental objective in the Water, Equatorial Lakes Region”. NELSAP also provides Sewerage and Sanitation sector is to support the funds and support for various transboundary projects overall national objective of providing clean and related to water and energy between Tanzania and safe drinking water to within easy reach, to satisfy its neighboring countries in the Nile River Basin,
46 including the Mara River Basin Management Project. Tanzania and outcomes from previously completed activities in Kenya have been incorporated into the process, following the mandate of this MoU. Memorandum of Understanding for Joint 3.1.2 Available Information Water Resources Management of the Data and information relevant to the EFA activities Transboundary Mara River Basin were collected from online sources and from the In 2015, the governments of the Republic of Kenya participating institutions. The data and information and the United Republic of Tanzania signed the were uploaded to central storage location during the “Memorandum of Understanding between the project where the team members could view, upload, Government of the Republic of Kenya and the and download files (documents, spreadsheets, photos, Government of the United Republic of Tanzania videos, spatial data, etc.). In this way, a database of for Joint Water Resources Management of the information related to the project was developed Transboundary Mara River Basin” (URT and that will be delivered to the LVBWB, the MoW, and Republic of Kenya, 2015). This document lays out NELSAP CU at the completion of the project for their the responsibilities of both countries when it comes records and continued use. to the dual management of the MRB, including the establishment of joint institutions for the “sustainable Available data resources utilized for this project development, management, and equitable utilization included: of water resources, including water allocation, water - Foundational documents for the project, including supply and sanitation, capacity building, data and the Nile Environmental Flow Framework, information information sharing, research and development.” on setting RQOs, and the Technical Offer for this This MoU is facilitated and supported by the LVBC. project. While environmental flows or the reserve are not - Hydrological data collected from the LVBWB and specifically stated in the MoU, it is a critical component analyzed by IHE Delft. of water allocation planning. When appropriate, the - Previous EFAs conducted in the Mara River from government of Kenya has been involved in the EFA in 2012 and 2017 and their associated reports and datasets (where available), as well as other EFAs conducted in Tanzania. - The Integrated Management Plan for the Mara Wetland (2018, developed by IHE Delft, Birdlife International, and WWF) and associated reports and datasets. - Information from a geomorphological study of the Mara Wetland conducted by IHE Delft, including drone footage, photographs, and spatial data. - Datasets and a literature review the MaMaSe project, a project that was completed in the Upper Mara River Basin in Kenya from 2014 – 2018. - National laws and guidelines for the United Republic of Tanzania related to water, wetlands, and environmental protection - Spatial data, reports, scientific papers, master’s theses and PhD dissertations related to the Lower MRB 3.1.3 Previous EFAs in the Mara River Basin The first phase of EFAs in the Mara River Basin was completed between 2006 and 2008 under the consortium by a team of experts from the United States, Kenya and Tanzanian universities, water management authorities in both countries, the Lake Victoria Basin Commission, and WWF among other stakeholders. The study aimed to determine the minimum flow levels required to maintain the reserve in the Mara River. Three sites were selected all located within MRB in Kenya. The findings of this study were adopted and summarized in a 2010 report by Lake Victoria Basin Council of Ministers. The recommendations from the 2006-2008 EFA called for additional studies on the Mara to provide more information on the status of the rivers, to assess the accuracy of the flow recommendations given at that time and suggest any necessary improvements. As a result, further studies were conducted from 2008 to 2010; low flow EFA sampling and long term
47 monitoring to address these recommendations. of the Mara. The plans considered management The low flow sampling focused on the physical and topics including i) water allocation and use; ii) water biological characteristics of the river at critical flow resources protection; iii) institutional development levels while the long term monitored water quality and and collaboration; iv) infrastructure development; v) discharge relationship as well as macroinvertebrate resource mobilization and financial management; and communities every two weeks. The findings from vi) livelihoods and entrepreneurship. Implementation these studies showed that the 2006-2008 EFA of these plans has been limited. recommendations were sufficient to maintain a healthy river ecosystem with the necessary control Water Allocation Planning Activities in the of the abstractions. The third phase building upon Mara River Basin phase two called for the extension of field sampling to Water allocation planning is an integral component Tanzania and this was undertaken between 2011 and of Integrated Water Resources Management and 2012. In this study, two sites in Tanzania were added Development plans being developed by Water Boards for assessment in addition to the three original sites in Tanzania and will be incorporated into the Lake in Kenya and flow recommendations given for each Victoria Basin Plan as it is developed. In addition, site. Reports for these earlier EFA studies outlining activities specific to water allocation planning are the methodology and results were developed and being, and have been, carried out in cooperation disseminated. with the efforts of the Water Board. Currently, NBI- NELSAP, the Sustainable Water Partnership, and The latest EFA assessment was conducted during the WWF are supporting the development of a WAP for MaMaSe project which was funded by the Embassy the Lower MRB. The reserve assessment reported of the Kingdom of the Netherlands in Nairobi, here is a component of this cooperation. Upon its Kenya from 2014 to 2018. The EFA was undertaken completion, efforts will be made to harmonize the in collaboration with the Kenyan Water Resources Tanzania Mara WAP with a similar plan that has Authority, Water Resources Users Associations in been developed and is undergoing modification Kenya, County governments (Bomet and Narok), for the Kenya part of the basin. The ambition is to the LVBWB in Tanzania, NBI, Lake Victoria Basin develop a single, transboundary WAP that can be Commission and a host of various stakeholders in the agreed by Tanzania and Kenya within the framework basin. Two field assessments were conducted between of the MoU for Joint Water Resources Management of November and December 2015 during the wet period the Transboundary Mara River Basin signed between and February and March 2016 representing the dry Kenya and Tanzania in September 2015. period. Seven sites were selected for assessment: 6 located in Kenya and one (Mara Mines) in Tanzania In 2013, the Lake Victoria Basin Commission also (same site where the 2006-2008 EFA was conducted). supported the development of a Mara River Basin- This EFA was part of the water allocation planning Wide Water Allocation Plan. That 5-year plan was effort for the Kenyan portion of the Mara River intended “to establish a reasonable and practical Basin, which included other demands in the basin framework for water allocation and water abstraction i.e. domestic, irrigation/agricultural, industrial, within the Mara River basin”. The authors of the livestock, wildlife etc. All these demands were used to plan noted that it was developed in the context calculate the water balance for the Mara River in the of major information constraints and should be upper part of the basin and subsequently the WAP, revised within 5 years. The information generated which is still in review by the relevant ministries in as part of this study and others supported by current Kenya. water allocation planning efforts is intended to fill information gaps and enable develop of a more Water Resources Activities in the Mara River detailed and well-informed WAP of the basin. Basin Water allocation planning activities began with Water Resources Planning Activities in the Mara the formation of the Transboundary Water for Catchment, Tanzania Biodiversity and Human Health in the Mara River Water resources planning in Tanzania is organized Basin Project that started in October 2005 and ended within nine major river basins, each administered by in September 2012. The project was a collaborative a Basin Water Board. The Mara River is a sub-basin effort under the Global Water for Sustainability of Tanzania’s Lake Victoria Basin and is administered program (GLOWS) with participation from several by the LVBWB. According to the Water Resources organizations including Florida International Management Act (2009), each Basin Water Board is to University (FIU), WWF Eastern and Southern Africa develop an Integrated Water Resources Management Regional Programme, World Vision, CARE Tanzania and Development Plan. The plan for the Lake Victoria and the Mara River Water Users Association. This Basin is currently under development and will include undertaking was funded by USAID and the adoption the Mara among other sub-basins. and implementation of findings done through the Lake Victoria Basin Commission. One of the Over the past decade, various activities and projects objectives of the project was to support governments in support of water resources planning in the Lower as well as local partners in Kenya and Tanzania to MRB have been conducted by the Lake Victoria Basin develop a water resources management plan for the Commission, NELSAP CU, WWF, and numerous Mara River Basin. other organizations. In 2014, NELSAP CU supported the development of Sub-Catchment Management Mara River Basin Management Project Plans for the Somoche and Tobora sub-catchments In the Mara River Basin, NELSAP operates the Mara
48 River Basin Management Project which began in 2006 assessment and implementation of the reserve is one and has the objectives of “improved water resources of a number of management actions to be guided by development through development multipurpose RQOs. Others include pollution prevention actions storage reservoirs for Irrigation, water supplies and regulations, and related actions controlling direct and small hydroelectric power, and improved river resource extractions such as fishing, sand mining, basin management through integrated watershed etc. In the case of the reserve, aspects of RQOs management projects”(NBI, 2015). It continuously related to meeting basic human needs for water and promotes the harmonization of policies and protecting ecosystems are most relevant. RQOs are management actions between Tanzania and Kenya, identified in Part VI of the Tanzanian Water Resource has completed feasibility studies for large-scale water Management Act (2009) as an instrument of water infrastructure projects (including Borenga Dam), and resource protection. The text of the Act calls for the has implementing small-scale water infrastructure MoW to “establish procedures which are designed projects. It also provides investments for water to satisfy the quality requirements of water users as quantity and quality monitoring networks, trainings far as is reasonably possible, without significantly for technical staff, and community outreach on altering the natural water quality characteristics environmental management issues and development of the water resource” [Part VI(a)32(2)(b)(ii)]. The options in the MRB. requirement to establish procedures for determining the reserve is also cited in Part VI of the Act, as the 3.1.5 Partnerships and Capacity Building two instruments are intended to work together in The EFA assessment was carried out in close water resource protection. collaboration with many local and international partners. NBI and GIZ provided the funding for the The articulation of RQO’s in the Act and link to the field work, guidance, and day to day liaison to the determination of the reserve is consistent with the project on the content and process of the assignment. NBI E-Flows Strategy (NBI, 2016a), which includes The field work was conducted by the EFA technical the setting of RQOs in the initial phases of e-flow team, staff from the LVBWB (Tanzania), the Water assessments. The RQOs are to be set in accordance Resources Authority (Kenya), IHE Delft (Netherlands), with local, national and regional governance (legal University of KwaZulu-Natal and Rhodes University and institutional). In the Mara River Basin, this (South Africa), University of Eldoret (Kenya) and refers to the Act as the legal basis and the LVBWB Sokoine University of Agriculture (Tanzania). The as the responsible institution. The detailed aspects of EFA technical team were also assisted by staff from RQOs are not specified by the Tanzania MoW but are the MoW, the Musoma District Fisheries Department, described in the technical manual accompanying the SENAPA, local government leaders from the wards NBI E-Flows Strategy, which have been adopted by the and villages, members of the water users associations United Republic of Tanzania through its membership (WUAs) and community members. in the NBI. The NBI manual adopts the description of RQOs specified in the 2004 South African National Capacity building was incorporated into all major Water Resource Management Strategy, which states activities in the project through hands-on learning that “resource quality includes water quantity and and interaction with experts. This project is the water quality, the character and condition of in- first time that RQOs have been developed for stream and riparian habitats, and the characteristics, a basin inside Tanzania as well as the first full condition and distribution of the aquatic biota : implementation of the NBI E-Flows Framework, Resource quality objectives will be defined for each which were described in detail during stakeholder significant resource to describe its quality at the workshops. During the field work campaigns, local desired level of protection” [3.1.1]. Reference to “the participants were encouraged to work with different desired level of protection” is important because it experts to learn about their field of study and gain acknowledges that RQOs will not be the same for hands-on experience conducting different field work all water bodies. RQOs may be set higher or lower methodologies. In particular, knowledge exchange depending on the needs of water users and legal between the water authority staff in Kenya and requirements for environmental protection. Tanzania was considered a high priority. Community members also provided the technical experts with RQOs are expressed as narrative statements of the information about the local conditions and recent desired quality aspects of the resource. For example, changes in ecosystem condition. The final flow setting the RQO related to water quantity may be “maintain workshop also discussed environmental flow science dry season low flows at levels sufficient to meet and implementation in detail, providing examples domestic and livestock needs but with only moderate from other projects around the world to improve the alternation of the ecosystem”. The advantage of knowledge of those who will be implementing the narrative statements is that they are more effective reserve in the Lower MRB. for communication, enabling stakeholders to better understand the stated objectives. Narrative 3.2 Resource Quality Objective statements are not, however, sufficient to guide water Setting resource managers because they are not measurable. RQOs are set to guide actions to protect water The manager must know “how much” flow during resources in the rivers, wetlands, aquifers, and lakes. the dry season is necessary to meet domestic and RQOs are intended to protect water and related livestock needs but with only moderate alternation aquatic biological resources at levels needed to meet of the ecosystem. This then requires measurable the needs of resource users and maintain ecosystems targets for “what is” moderate alternation of the in a desired environmental management class. The ecosystem. Therefore, narrative statements must be 49 accompanied by measurable targets that the resource of the Tigithe River and flows into the Mara Wetland. manager can set and monitor to check that objectives The Lower Tigithe (Tigithe Chini) WUA is established are being met. in this RU. 3.2.1 Resource Units North Mara: This RU encompasses the area on Spatial areas needed to be determined for which the north side of the Mara Wetland, approximately RQOs were developed. They were selected to align between the inflow of the Tigithe River and the outlet both with existing management structures, in to Lake Victoria. The North Mara (Mara Kaskazini) particular the WUAs, as well as the biophysical study WUA is established in this RU. sites selected to study flow-ecology relationships. In total, seven resource units (RUs) were delineated for South Mara: This RU includes the area to the south the Lower MRB (Figure 3 3). Working upstream to of the Mara Wetland, approximately between the downstream, they are: southern tributaries to the wetland and the outlet to Lake Victoria. The South Mara (Mara Kusini) WUA is Serengeti: This RU includes SENAPA and many of established in the RU, although only within the area the tributaries that flow into that area. There is no adjacent to the wetland boundary. WUA established in this region, however the RQO process is being conducted in close collaboration There is an additional area labelled Mara Mines in with SENAPA technical and management staff. the central part of the river basin that is not included in an RU. This section does not have an established Tobora: This RU follows the sub-basin boundary for WUA or management authority and was not assessed the Tobora River, which flows into the mainstem of in this RQO workshop. However, RQO statements the Mara River downstream of SENAPA. The Tobora were developed during the EFA process. WUA is established in this RU. 3.2.2 Stakeholder Workshop Somoche: This RU follows the sub-basin boundary An RQO stakeholder workshop was held on November of the Somoche River, which flows into the mainstem 7th and 8th, 2018 in Tarime, Tanzania. The objective of the Mara River downstream of the Tobora River. of the workshop was to develop narrative RQOs for The Somoche WUA is established in this RU. the RUs of the Lower MRB with local stakeholders. Participants in the workshop were guided to this Upper Tigithe: This RU is the upstream portion of objective by working through a series of activities the Tigithe River and flows into the Lower Tigithe RU. designed to develop understanding of the process, The Upper Tigithe (Tigithe Juu) WUA is established gather needed information, and articulate the in this RU. narrative RQOs. It was also a chance to encourage knowledge exchange and build relationships between Lower Tigithe: This RU is the downstream portion different types of stakeholders in the basin through
FIGURE 3 3: RESOURCE UNITS FOR THE LOWER MRB
These terms are defined as the following: 3 Quantity: Including pattern, timing, water level and assurance of instream flow. Quality: Including physical, chemical and biological character of the water. Habitat: Including character and condition of the instream and riparian habitat. Biota: Including character, condition and distribution of the aquatic biota 50 small group discussions. All workshop materials and data gap is present in the 1990’s, and within the 49 results can be found in the full RQO workshop report years of data 30 percent is missing. found in Annex A. The RQO Stakeholder Workshop was attended by 57 3.3.1 Data Regionalization participants from 19 organizations, including local, Because of the limited availability of long-term regional, and national stakeholders, project partners, hydro-meteorological data sets in the catchment, the and support staff. The stakeholder organizations reconstruction of river flow data was based on earlier included all six WUAs, the Mwanza and Musoma works carried out under the MaMaSe and Sustainable offices of the LVBWB, the Tanzanian MoW, the Water Partnership projects. A detailed description of Zonal Irrigation Office in Mwanza, SENAPA, the methodology can be found in the reports MaMaSe, Tanzanian universities, and local and national non- 2017 and SWP, 2019 , and here only a summary of governmental organizations. Partner organizations the methodology is given. The reports describe the included IHE Delft, NBI-NELSAP, WWF-Tanzania, methodology for a water availability assessment the Sustainable Water Partnership, USAID Kenya, using long-term historical data sets of precipitation and the Stockholm Environment Institute. and discharge. Objectives of these assessments were (i) to regionalize average monthly and average annual The participants were split into smaller groups, one discharge data, (ii) estimate flow duration curves, group for each RU. The participants were asked to (iii) to setup long-term water balances for the sub- place themselves in a group where they lived in the catchments within the Lower MRB, and (iv) to assess RU, were familiar with the conditions inside the RU, changes in the hydro-meteorological time series data or had a professional or personal connection with the sets (SWP, 2019). RU. Each group contained 4 to 7 participants, with two representatives from the WUA residing within the RU Methods used for reconstruction of river flow series: and a mix of participants from other organizations and government institutions. The groups were 1. Catchment delineation based on SRTM 90 intentionally mixed to provide a combination of meter data using the Pfafstetter coding system. local knowledge, academic understanding, and professional decision-making. Each group was asked 2. Monthly precipitation time series of 25 stations in to complete four activities: and around the Mara River Basin. - Assess the impacts on resource qualities from 3. Areal precipitation estimation using the Thiessen external pressures. polygon method. - Assess the importance of resource qualities to water 4. Monthly discharge time series of four gauging users. stations. - Assess the fitness for use of resource qualities. 5. Regionalization of discharge values using a runoff coefficient approach. - Develop draft RQO statements. 6. Filling of missing data using cross-correlation The draft RQOs were then reviewed and revised by between the two neighbouring stations. the EFA technical team. Final RQO statements can be found in Section 5.1. 3.3.2 Long-term Annual Water Balance Results of the regionalized long-term annual water 3.3 Hydrologic Foundation balances for the EFA sites are given in Table 3 3. The following is a summary of hydrological analyses Annual precipitation values range from 936 to 1,100 completed in the hydrology starter document for mm/yr, and evaporation values from 860 to 1,018 the Lower Mara EFA (Annex C) as well as the mm/yr. The Tigithe EFA catchment shows the highest hydrological analysis completed in the Water precipitation and evaporation values, whereas the Availability Assessment (SWP, 2019) for the water Kogatende EFA catchment has the lowest values. This allocation planning activities in the basin. For full can be explained by the fact that the upstream area details, please refer to these reports. of Kogatende has the highest relative contribution of the dryer and lower yielding sub-catchments of Talek The hydro-meteorological observation network in and Sand (located in Kenya) in comparison to the the Lower MRB is limited. In total, 13 stations are more downstream EFA sites. Regarding the runoff located in the Lower MRB: four automatic weather contributions, most of the EFA sites have comparable stations, five rainfall stations, and four hydrometric values around 70 to 80 mm/yr. One exception is the stations. In June 2018, only one rainfall station was fully operational and three of the hydrometric stations were partially operational (SWP, 2018). Available flow data are limited to the Mara Mines site with data being available from 1969 to 2018. A large
51 FIGURE 3 4: METHODOLOGY FLOW CHART FOR THE WATER AVAILABILITY ASSESSMENT IN THE LOWER MRB (SWP, 2019)
TABLE 3 3: CATCHMENT CHARACTERISTICS OF THE EFA SUB-CATCHMENTS 52 Somoche EFA catchment with a value of 48 mm/yr.
for average monthly discharge are similar for the EFA 3.3.3 Regionalized Monthly and Annual Flow sites located on the mainstem Mara River and the Values wetland, while the three tributaries follow a similar In Figure 3 5 a summary of the regionalized flow pattern but much smaller in magnitude. The runoff values for the EFA sites is presented. The hydrographs
FIGURE 3 5: AVERAGE MONTHLY DISCHARGE (LEFT) AND RUNOFF REGIMES (RIGHT) AT THE EFA SITES regimes are also fairly similar across the EFA sites, precipitation input data gap filled daily time series with Somoche having the highest runoff ratio. The of available stations were used. After extracting the numerical flow values can be found in Annex C. annual maximum flow values from the time series 3.3.4 Frequency Analysis a Gumbel distribution (Generalized Extreme Value A frequency analysis was carried out for annual distribution Type-I) was used and fitted to the data maximum daily flow values at each EFA site. Daily sets. Results of the frequency analysis for the EFA time series for the EFA sites were generated using sites can be found in Table 3 4 and Annex C. a linear reservoir based rainfall-runoff model. As Although the model was calibrated against measured
TABLE 3 4: SUMMARY OF FREQUENCY ANALYSIS FOR MAXIMUM DAILY FLOW VALUES station data, this approach contains considerable found in the report (SWP, 2019). None of the seven uncertainties, and should only be seen as a first considered precipitation stations in the Lower MRB attempt to estimate return periods of annual maxima. showed a significant positive or negative trend in the 3.3.5 Trend Analysis annual precipitation amounts. Autocorrelation was A trend analysis of hydro-meteorological parameters detected for the station 9134027 Lolgorien, and the within the Lower MRB was carried out using the calculated pre-whitened trend was not significant. DScreen software tool (Venneker, 2011) and the For the discharge trend analysis the average annual Indicators of Hydrologic Alteration package (Richter (MQyear) as well as the maximum (HQmonth) and et al., 1996; The Nature Conservancy, 2009). A minimum (NQmonth) average monthly discharge detailed description of the methodology can be values were selected for the station 5H2 Mara Mines. 53 Autocorrelation was detected for the minimum parts of the Pangani and Wami-Ruvu Basins as well monthly discharge values and therefore a pre- as along the eastern and western shores of Lakes whitened trend was calculated in this case. The time Victoria and Tanganyika. Defining environmental series of 5H2 Mara Mines show diverse trends, but characteristics include moderate elevation and none of them is statistically significant. The daily slope; moderate to low (variable) precipitation with discharge values of the station Mara Mines also low (variable) seasonality; lowest subsoil porosity, showed no statistically significant trends. and high vegetation cover and low agricultural land cover. 3.3.6 Hydrological Assessment of EFA Study While this classification allows the rivers and Sites streams in the Lower MRB to be compared to other Using the regionalization approach, hydrological rivers nationally, it is not at a fine enough scale to analyses were conducted for each EFA study site. show differences within the basin. It is important Summarized information on land use, catchment to recognize the differences in the river attributes size, and discharge information can be found for during an EFA, even in a generalized way, during the each site in Section 4. More detailed information can selection process to determine EFA study sites. This be found in the full hydrology starter document in helps to ensure that different types of river systems Annex C. are analyzed and incorporated into the study, particularly since the management recommendations Ecosystem Classification often change due to the natural river types. Since There is no official ecosystem classification map RUs were already defined in the RQO process, or methodology defined for Tanzania. As part of these boundaries were used to provide general previous environmental flow work, a national river classifications based on known information and classification methodology was proposed and carried experience working in the lower MRB. During the site out using available information on 20 different selection process, the technical team generalized the environmental attributes, including average channel hydrological characteristics of the RU to ensure each slope, catchment area, elevation, annual rainfall, were captured. In general, there are three distinct and soil porosity, among others (USAID, 2018). The hydrological areas: the mainstem of the Mara River methodology was applied in detail to the Rufiji River with perennial flow, tributaries which have very Basin in central Tanzania, and then applied to all nine low base flows and high flows are driven by rainfall river basins in Tanzania at a coarser level (Figure 3 events, and the wetland which acts as a gradient 6). between the influence of Lake Victoria downstream According to this analysis, almost the entire Lower and the Mara River upstream (Figure 3 7). Additional MRB is classified as Class H, which is defined as: biological and geomorphological attributes were also his river class is found in every major Tanzanian discussed, including habitat function, approximate river basin, with major concentrations in the central slope, and substrate. In general, these also align
FIGURE 3 6: PREDICTED RIVER CLASSES OF ALL RIVER SEGMENTS IN TANZANIA ACCORDING TO THE DEDUCTIVE CLASSIFICATION (USAID, 2018). 54 FIGURE 3 7: GENERAL ECOSYSTEM CLASSIFICATION OF THE RESOURCE UNITS IN THE LOWER MRB
TABLE 3 5: ECOSYSTEM CLASSIFICATION DETAILS FOR EACH RESOURCE UNIT AND ASSOCIATED EFA STUDY SITE 55 with the hydrological characteristics, maintaining ecological and human communities can be complex, three distinct classifications. Details on all ecological and often require detailed scientific studies to classification attributes can be found in Table 3 5. understand how these relationships change over 3.5 Flow Alterations a hydrologic year and over many years. In high Flow alterations are a measure of the deviation of conservation value rivers like the Mara, it is advised current-condition flows from baseline- (or natural-) to use a holistic EFA methodology to help determine condition flows. Significant flow alterations occur these linkages. In this study of the Lower Mara in river basins regulated by large infrastructure or River we have applied a modified Building Block with high water demand relative to water availability. Methodology (BBM, King, Tharme and Villiers, In the Mara, there are currently no engineered 2008), which is consistent with the methodology structures that significantly alter flows. Thus, applied in the Upper Mara River in Kenya. This under most flow conditions the flow regime is near methodology, developed in South Africa and cited natural. During periods of low flow, direct river water in the Nile E-Flows Framework, combines existing withdrawals for domestic and agricultural use may scientific literature, detailed field studies, and the have a measurable effect on flow levels, but no data knowledge of a team of experts to determine flow- are available and the magnitude of the effect is likely ecology and social relationships. These are then minimal in most reaches of the river system. Mango applied to set environmental flow levels that meet the et al., 2011 modelled the effect of land use change on RQOs. hydrological runoff in the Nyangores tributary of the Upper Mara in Kenya, and their results did suggest The BBM assesses flow-ecology ecosystem services a small decrease in dry-season flows and increase linkages for different components of the hydrograph in flood flows, but the results remain to be verified (or “building blocks”, Figure 3 8). These include low by flow data and cannot be directly extrapolated to flows, small to medium floods, and large floods. conditions in the Lower MRB in Tanzania. In this The flow requirements of these building blocks study we have therefore not considered current flows are determined using physical and biological to deviate significantly from baseline or natural requirements of fish, macroinvertebrates, and flows. The ramifications of this consideration is riparian vegetation (flow-ecology relationships) as that observed degradation in ecological condition well as the use of the ecosystems by local communities of the river system is assumed to be caused by (ecosystem services). Specific indicator(s) for each pressures other than flow alterations, which may component are determined by specialists based include contaminant discharges and direct human on available data and professional judgement. The interventions (like over-fishing or clearing of riparian indicator(s) selected then determine the specific vegetation). data collection methodologies used in the field. The information for each component (hydrology, 3.6 Flow-Ecology Ecosystem hydraulics, water quality, geomorphology, riparian Services Linkages vegetation, fish, macroinvertebrates, and social uses) The relationships between flow and dependent are combined in technical documents (called starter
FIGURE 3 8: SCHEMATIC ILLUSTRATION OF THE MAIN COMPONENTS OF A GENERIC FLOW REGIME (COMPRISED OF INDIVIDUAL FLOW BUILDING BLOCKS) CONSIDERED IN THE BBM (FROM CDM SMITH, 2016).
56 documents), that serve as the technical inputs to section area, mean and maximum depth, width, the flow setting technical meeting, where detailed mean and maximum velocity, and the water environmental flow levels are set. temperature. For discharge measurements using the In addition to collecting physical data, the BBM calls SonTek River Surveyor M9, final discharge values for collection of social data to assess how communities were determined using the River Surveyor Live make use of ecosystem services provided by the river, software. All measurements of the River Surveyor its riparian corridor, and related wetlands. Emphasis at one cross-section were checked regarding their is then devoted to protection of underlying ecological quality, and before calculation an average doubtful processes providing the most valued services. measurements were excluded from the analysis. In Detailed descriptions of the different BBM case of the OTT Acoustic Digital Current meter and components are described in the following sub- the SonTek Flow Tracker, the Velocity-Area method sections. was used to determine discharges. Sampling verticals were spaced 0.5 – 1m, with velocity measurements 3.6.1 Flow-Ecology Components being done at 0.6 of the depth at each vertical. To Social Uses determine the average discharge, the Mid-Section The objective of the social study was to provide and Mean-Section method was used. information from the perspective of the community members on how they utilize riverine resources for Measurement of the cross-sections were carried sustaining their livelihoods. This could be either out using the real-time kinematic global navigation in terms of food, crop farming, source of building satellite positioning system EMLID Reach RS/RS+ material, medicinal value, fuel and other purposes. (Emlid, 2019). The first receiver was installed on a The study also identified key issues of concern fixed tripod as the base station, and a second receiver raised by the community members relating to these was used as a mobile rover to determine the locations resources and what improvements they may wish and elevations of the cross-section points. The to see in order to promote a healthy ecosystem that receivers were used in differential mode to correct will be beneficial not only to the ecology but to them the cross-section point positions relative to the base as well. It also provided an opportunity to learn station and to obtain centimetre accuracy. from the community some of the historical changes that may have happened to the resource in terms In order to reconstruct the historical flow regime, data of its availability, quantity or quality and making regionalization method was used because of limited a comparison to what is currently the situation. availability of long term meteorological data for the Participatory Rural Appraisal was the approach sites. This is the process of using known hydrologic employed by the social scientist for data collection characteristics (catchment size, areal rainfall, while Focus Group Discussion was the technique evaporation) of a particular catchment/ hydrological selected for this purpose. unit to calculate similar variables in an area with no/ insufficient data. See Section 3.3 for details. Hydrology Hydrological analysis of the EFA sites was done Hydraulics in order to determine the flow regime of the Mara The aim of the hydraulic assessment is to extrapolate River over several timescales. The regime reflects the and translate flows into stage and velocity data for annual variability, timing and seasonal distribution the various sites. This is a crucial link between the of flows and the extent to which these flows keep re- hydrology/water allocation and hydraulic habitat/ occurring (return period). Hydrological assessment channel maintenance. is important as it impacts on the ecological and geomorphological processes such as shape and size A single transect for hydraulic observations of the river channel and behavior of the aquatic and modelling was selected per site based on its organisms. The analysis was done by combining importance as critical habitat for organisms and field observations (on-site discharge measurements), sensitivity to low flows. Transects were then marked literature studies of existing technical reports and and 50m tape measures or ropes were used to guide analysis of available precipitation and discharge time surveying. Data gathering involved surveying the series. topographic data along the transect (perpendicular to flow); survey of water levels and measurement of Discharge measurements at the EFA sites during the depth and velocity along each transect. field campaign were carried out using three different devices depending on the flow situation and cross- Land based surveying was done with survey section characteristics: grade equipment (Topcon Total Station or EMLID Differential GPS). For sites with deeper water with - ADCP = SonTek River Surveyor M9; Acoustic potential wildlife dangers, a SonTek River Surveyor Doppler Current Profiler. M9/S5 using acoustic Doppler technology was used to determine depth and velocity at a large number (>100) - OTT ADC = OTT Acoustic Digital Current meter. of verticals along each transect. For very shallow depths where the River Surveyor could not capture - Flow Tracker = SonTek Flow Tracker handheld meaningful data, a handheld acoustic Doppler OTT Acoustic Doppler Velocimeter. ADC was used to capture flow velocity and depth. For measuring the discharge, the channel was divided into Observed parameters included discharge, cross- 20 verticals to capture depth and flow velocity data.
57 The observed hydraulic data were used to develop flow requirements and flow related impacts. Other and calibrate hydraulic models using Hydrologic zones were floodplain and macro-channel bank. All Engineering Centre’s River Analysis System (HEC encountered plants were recorded and subsequently RAS) software. Frequency distributions of depth- identified i.e. family, genus and species in the field velocity classes were calculated using Habitat Flow (whenever possible) and confirmed at the National Simulation Software (HABFLO). Herbarium of Tanzania in Arusha where all voucher specimens are deposited. Estimation of aerial cover Geomorphology (in percent) was done using visual observation. The shape of the river channel results from fluvial Information on extent, period and duration of processes of erosion, transport and deposition. inundation was collected through interviews. Geomorphic features in the channel play a significant role in determining the availability and diversity of Fish physical habitat and will influence the nature of the Fish can be used to reflect the combined effects of aquatic ecosystem. environmental changes that have happened and can be used to reflect the environmental health of the Fieldwork methods included in-depth channel river. The presence of a large diversity of fish species surveys, more general reach descriptions and and abundance can help in understanding the broader landscape assessments. A single river cross functioning of a river. Knowledge of the conditions section was surveyed per site with a Total Station or needed by fish for spawning, hatching, growth Differential GPS for terrestrial sections and a SonTek etc. can be used as a guide for recommending flow hydro surveyor for the portions of deep and fast flow requirements. Fish species were sampled to represent (for crocodile infested water). Georeferenced land the species composition and proportional abundance based photos were taken and sketches were made of of the assemblage for each site. Fish were primarily geomorphic features and their sediment composition. sampled by electrofishing different Geomorphic Sediment from the riverbanks, inset benches and Habitat Units (GHU). Cast netting, fyke nets and river bed were collected using a 30 mm diameter hand seine netting were also undertaken in suitable GHUs, corer to a depth of 50 mm. Five samples were taken with species diversity and abundances recorded for per feature and composited to form a representative respective efforts. The catches of local fishermen sample. A subsample of the composite sample was were also considered for the study, but only for the used for particle sizing. Particle size was determined Mara Wetland system. Fish were identified in the using a Malvern Mastersizer 3000 (for silt and clay field, photographed, measured (standard length) and samples), the Eijkelkamp Sand Ruler (for sand sized released back into the respective systems. particles) or tape measure (for larger particles). For the larger particles, a sample of 100 randomly selected The Habitat Cover Rating (HCR) method described clasts were measured with a tape measure along the by Kleynhans, 1999 was adapted and implemented b-axis to determine the grain size distribution. for the study to characterize habitat for each survey site. The HCR was calculated according to the rating River reaches upstream and downstream of the cross of the relative contribution of various velocity-depth section were explored to develop an understanding classes, where 1 = Rare/poor (<5 percent), 2 = Sparse/ of river character, geomorphic habitat template and poor (5-25 percent), 3 = Moderate (25-75 percent), key sediment processes taking place. Catchment- and 4 = Extensive (>75 percent). An overview of the wide landscape connectivity and erosion extent and velocity depth classes described by the HCR approach severity was assessed visually while driving to and is presented in Table 3 6. between sites. Satellite images in Google Earth were used to explore and qualitatively assess areas of the Cover features are rated within each depth-flow catchment that was difficult to access by road in the class using the same scale in order to calculate the given time. HCR. The cover features were summed for each of the depth-flow classes. The HCR at each site was Riparian Vegetation then calculated based on the contribution of each Riparian zones connect terrestrial and aquatic depth-flow class multiplied by the summed cover ecosystems and are important in providing food to feature ratings for each depth-flow class. To complete instream organisms, soil conservation and regulating linear redundancy analysis and to determine the water temperature from their canopy cover among relationships between habitat ratings and fish species other uses. Different riparian plants are adapted to different flows and can be used to indicate high and low flows. Description of surveyed sites was made to highlight key landscape information including dominant vegetation types and site conditions e.g. highly degraded, moderately degraded, slightly degraded and not degraded. Plan view and sketches of cross sections were drawn and photographs taken at each site. Along the set transects, riparian vegetation zones and sub-zones (marginal, lower, upper) were both identified. Sub-zones were also identified because species composition and distribution differ in different sub-zones with implications on TABLE 3 6: THE HCR CLASSIFICATIONS FOR RESPECTIVE FLOW VELOCITY AND RECORDED DEPTHS 58 Canoco version 4.5 (Braak and Smilauer, 2002) was were preserved in formalin in separate containers used. Species abundance was assessed against depth and taken to the laboratory for further processing class and cover feature ratings. and enumeration of abundances of the various taxa. Macroinvertebrates These data were particularly useful for statistical Macroinvertebrates are good indicators of the analyses to determine the preferences of the various ecological health of the river because of their taxa in terms of flow velocity, depth and substrate sensitivity to river flow alterations and water quality. type. The South African Scoring System version 5 Therefore their variation in terms of numbers and (SASS5), and the Tanzania River Scoring System species occurrences can be used to explain the (TARISS) biotic indices were used for the assessment biological and/or ecological changes in the river of the present ecological status of the sampled sites. ecosystem. The EPT index was also applied, which compares the amount of specific taxa sensitive to water quality During field sampling, a modified kick net with (Ephemeroptera (mayflies), Plecoptera (stoneflies), a mesh size of 1,000 μm was used to sample the and Tricoptera (caddisflies)) to the total number of macroinvertebrates within a prescribed time limit taxa found. and/or areal coverage. The stones-inside-current and bedrock was searched (‘kicked’) for a period of Water Quality 2 to 5 minutes. Similarly, stones-out-of-current and Water quality refers to the physical, biological bedrock were searched for 1 minute. and chemical characteristics of a water body that determines its suitability for domestic as well as The SIC and SOOC samples were combined into ecological purposes. There are no water quality a ‘Stones’ sample. Suitable stretches covering two requirements for ecosystems in Tanzania, and as meter marginal vegetation was swept as well as such, there will be no comparisons made between the aquatic vegetation covering one square meter. This field results and any water quality standards. There, represented the ‘Vegetation’ sample. Gravel, sand and however, related national standards, international mud sample was stirred and swept for one minute guidelines, and general ecological requirements that and filtered to check for presence of any macro- help to put the water quality field results into context. invertebrates. Hand picking and visual observation was also employed for 1 minute and biotopes where The Tanzanian National Bureau of Standards macro-invertebrates were found recorded in a score developed national standards for drinking water from sheet. Loose stones were picked and screened for the public water supplies (Table 3 7) and for wastewater presence of benthos. The South African sensitivity discharge from municipal and industrial sources score (SASS) and the average score per taxon (ASPT) (Table 3 8), which may provide some reference for were also used to characterize macroinvertebrates values obtained in the field. These should not be taken at each site. In addition to the invertebrate sample, as standards that the water quality measurements water depth and substrate type data were collected in the field should adhere to since these values were for each habitat. All samples from the three habitats developed for water that has undergone human
TABLE 3 7: SELECTED PARAMETERS FOR TANZANIA DRINKING WATER STANDARDS
59 TABLE 3 8: SELECTED PARAMETERS FOR TANZANIA MUNICIPAL AND INDUSTRIAL WASTEWATERS treatment and were not developed for natural category 1 being the highest integrity and category 4 conditions. being the lowest. These benchmarks are just examples and do not explicitly apply to Tanzania since the The United Nations Environment Programme and the intent is that Tanzanian government would develop United Nations University Institute for Environment their own benchmarks that are more representative and Human Security also developed a policy-driven and suitable to conditions in the country. process on how countries can develop their own Fish and macroinvertebrates have varying degree water quality guidelines for ecosystems (UNEP/ of tolerance and are therefore considered as UNU-EHS, 2016). The intent of this document is that good bioindicators of water quality. Whereas it provides a framework for countries to develop their macroinvertebrates are normally used for short own water quality guidelines for ecosystems, but it term indication of water quality, fishes can be used does provide physico-chemical benchmarks that to show the long term variation of the same. The would be considered at both ends of the ecosystem physicochemical properties of water affects not condition spectrum (“high integrity” and “extreme only the health of aquatic ecosystems but also its impairment”, Table 3 9). The recommendation is functioning. Most specifically, pH levels, temperature, that spectrum be separated into four categories, with dissolved oxygen, specific conductivity, nutrients and
60 TABLE 3 9: EXAMPLE PHYSICO-CHEMICAL BENCHMARKS FOR FRESHWATER ECOSYSTEMS (UNEP/UNU-EHS, 2016)
61 discharge are the main parameters that have a strong oxygen (DO) measured using the WTW HACH effect on the health of aquatic environment (Boney, meters. 1989). - Non-toxic constituents: Electrical conductivity According to studies, low pH and dissolved oxygen (EC) using a WTW meter and turbidity measured levels (Dallas, 2008), high temperatures (Hayes with a clarity tube and Young, 2001), increased sedimentation (Koehn, - Nutrients: Nitrates (NO3-), nitrites (NO2-) using O’Connor and Research, 1990) and decreased flows the nitrate strip are some of the changes that may negatively affect the The laboratory samples were collected in 2 * 25 ml abundance, population, distribution and diversity plastic bottles after filtering using a 0.45 micron of aquatic biota. While some aquatic organisms are filter except the sample for Total Arsenic which adapted to specific conditions of the water and a was unfiltered. Samples collected for laboratory variation may alter how they function (Jackson, 1997), measurement included Ammonium (NH4+), total others are able to adapt to the changes over time. For organic carbon and total nitrogen. For the total example, macroinvertebrates such as Ephemeroptera organic carbon, NH4+, and total nitrogen samples, (Mayflies) and Trichoptera (caddisflies) are preservation was done using two drops of 0.2M intolerant to low DO while Mollusca (molluscs) and sulphuric acid while for Arsenic, Nitric acid was used. Chironomidae are quite tolerant to low levels of DO These were analysed at the laboratory at IHE Delft (Connolly, Crossland and Pearson, 2004). following standardized laboratory procedures. For the functioning of the aquatic organisms, an 3.6.2 EFA Study Sites and Field Campaigns optimal level of the chemical parameters has to To identify the key linkages between flow, ecology, be maintained. For example, according to studies and ecosystem services, two biophysical and one conducted by (Camargo, Alonso and De La Puente, social field campaigns were completed. Study sites 2004), 2 mg/L of nitrate nitrogen is ideal to ensure were planned to align with the RUs developed during the conservation of the most sensitive freshwater the RQO process, with two villages selected within species. Dissolved oxygen level ranging between 4-6 each RU for the social survey and one biophysical site mg/L and BOD levels > 5 mg/L (Bora and Goswami, selected within each RU (Upper and Lower Tigithe 2017) are necessary for creating an ideal environment RUs share one biophysical site). Figure 3 9 provides for aquatic microorganisms. Water quality involved a map of the study sites and RUs and Table 3 10 in-situ measurements as well as sample collection for assigns an EFA study site name (which will be used lab analysis. Parameters measured on site included: in the remainder of this document) and shows how the different study sites correspond geographically - System variables: pH, temperature and dissolved to each other. The details of each field campaign are
FIGURE 3 9: MAP OF THE STUDY SITES FOR THE SOCIAL SURVEY, BIOPHYSICAL FIELD CAMPAIGNS, AND RUS 62 TABLE 3 10: LIST OF THE CORRESPONDING EFA STUDY SITE NAMES AND RUS provided in the sections below. villages located far from the river/wetland. - Population: The relationship between resource quality and population is evident, particularly in the Social Survey dimensions of water quality, habitat and biota. The The survey was conducted in seven centers which had higher the number of water users the more likelihood been selected prior to the actual day of data collection. the resource will be destructed. For this case, villages The centers include Somoche, Upper Tigithe, Lower with high population were preferred for selection. Tigithe, Mara North, Mara South, Kogatende and The total number of participants in each village was Tobora. From each of the seven centers, two villages about 40 who were selected using the village register were sampled hence a total of 14 villages (Figure 3 with the assistance of the village leader. The selection 9). Selection of the villages was made based on the was random but the gender and age of the participants following considerations: were taken into consideration. During the Focus - Accessibility: Three levels were created to measure Group Discussions, the participants were divided accessibility to the study villages and coded using the in three groups of 13 people with each group being numbers 1 to 3. 1 represented a village that is hardly facilitated by an expert. Each group had a different accessible, 2; fairly accessible and 3; easily accessible. discussion theme i.e. general village profiles, social The scoring values were based on the knowledge and economic issues, natural resources available and and experience of the researcher and key informant their environment, and biophysical analysis of water interviews. The key informants for this case were the bodies in each surveyed village. All the information local government officials whom the discussion was gathered was clearly written on flip charts which made through telephone/skype conferences. were dated and numbered and supported by audio - Diversity of economic activities: Economic activities recordings to enhance data transcription when varied slightly depending on the dominant social reporting. groups. In areas where mixed social groups were dominant, more diverse economic activities were 3.6.2 1st Biophysical Field Campaign practiced and the impact to the quality of the resource The purpose of the biophysical field campaigns was to predicted, especially when there was limited control allow the hydrological and ecology experts to collect mechanism. In this case, the 1 to 3 code was also used field data from around the Lower MRB. Seven sites representing villages with low, moderate and high were considered for the biophysical assessment: three diversity of economic activities respectively. sites were selected close to the outlets of the major - Proximity to the preferred river/wetland: The tributaries on the Mara River (Somoche, Tigithe and location of a household with reference to the resource Tobora), two sites along the mainstem of the Mara determines how the resource will be used. In this River (Kogatende and Mara Mines) and two sites case, villages close to the resource was selected in in the Mara Wetland (Bisarwi and Mara Wetland) some RUs while in others, the preference was for (Figure 3 9). Location (latitude and longitude) and channel slope details can be found in Table 3 11 and
TABLE 3 11: LOCATION DETAILS FOR THE BIOPHYSICAL STUDY SITES 63 TABLE 3 12: BIOPHYSICAL STUDY SITE DESCRIPTION (TRANSECT, LOCATION AND FLOW) site descriptions in Table 3 12. the MaMaSe EFA in 2015 and also by GLOWS- FIU in 2012 (GLOWS-FIU, 2012). These two sites were therefore considered for assessment in order Selection of sites for the biophysical survey was based to allow comparison of past and current flow on identifying a river section that had a range of recommendations and also as a means of building environmental conditions characteristic of the whole information on the same site. river section. The following factors were considered: The first biophysical field campaign was conducted - Ease of accessibility to the site: Sites that could be when the water level was at its lowest (according to easily accessed by vehicles were mostly preferred as basin hydrograph) and was completed from the 4th this reduced the time taken to walk to the site. A lot to 9th of February 2019. Final analyses of the data of heavy equipment was also carried and used by the collected in the field were included in the individual team in collecting data and therefore this minimized starter documents (Annex B through Annex I). the need to carry them over long distances. - Diversity of physical habitat: A good site was one 3.6.2.3 2nd Biophysical Field Campaign that exhibited a range of diversity in terms of flow The second biophysical field campaign was intended sensitive sections with pools and ripples important to be completed during periods of high flow (according for aquatic and riparian species as well as for to the basin hydrograph). The rainy season preceding geomorphological analysis. this field campaign was delayed and greatly reduced, - A site that did not exhibit signs of being overly resulting in similar conditions to first biophysical modified: Part of the process of selecting a suitable field campaign. This is not ideal since the technical site involved assessing one that had minimal human team was not able to assess the conditions at higher interference. An overly modified river section would flows, but they were able to view the system response have a bigger effect on the number and species during a prolonged low flow period, finding a diversity (aquatic and riparian) that can be found and surprising resilience in the system to such conditions. hence may not be a good representative for the river However, it should be noted that the previous year stretch. was exceptionally wet and there was at least one - Proximity to an existing monitoring site: Some of the freshet that moved through the system between field sites selected e.g. at Kogatende and Mara Mines were campaigns. located close to existing flow gauging stations. This was useful because some of these stations already The second field campaign was conducted from the had some historical flow data and hence provided an 21st to 24th of May 2019, with data collected for opportunity to compare past and present results. The discharge, hydraulics, fish and water quality. The rating curves developed for these sites would also methodologies used were the same as in the first field be useful for the LVBWB officers for monitoring the campaign. river levels. - Safety factor: This was a major concern especially 3.6.3 Starter Documents when sampling in the SENAPA site (Kogatende) due One of the critical outputs of the BBM is the to the presence of hippos, crocodiles and other wild development of a starter document for each specialist animals. As a result, it was necessary to hire armed component. For this effort, starter documents were guards to keep a watchful eye on possible dangers developed for water quality, geomorphology, fish, and warn the team whenever a wild animal was macroinvertebrates, riparian vegetation, and social spotted nearby. uses. Each starter document contains information - Alignment with previous EFA sites: Out of the that is critical for understanding the flow-ecology- seven sites selected, only two sites; Kogatende and ecosystem services linkages at each site and Mara Mines had previously been surveyed during within the system as a whole. To achieve this, each 64 starter document follows the same structure so the recover from any interventions. information for each specialist component can be more However, the Tanzanian government has defined its easily combined and analysed. The structure included own 3 classes of PES that has been used to classify providing detailed site descriptions, determining a set the sites. To keep the process as country-specific as of site metrics, deciding the important indicators and possible, these following classifications were used: their associated management objectives, and then Quantity Class A: Near-natural: The natural flow determining the required physical habitat conditions regime is to be retained for their indicators (such as water depth, velocity, Quantity Class B: Somewhat altered: The natural inundation, and/or water quality as parameters). Also flow regime is affected by water withdrawals, important to this effort is the inclusion of confidence impoundments and/or discharges, but the critical rankings and data gaps to show where there are data aspects of the flow regime are retained so that effects inadequacies and how these may have impacted the on the ecosystem are small. analysis by the technical team. Brief descriptions of Quantity Class C: Significantly altered: The River is each part of the structure have been included below. affected by water withdrawals, impoundments and/ Summaries of the study site results are provided in or discharges to the degree that at least one aspect Section 4-14 while the full starter documents are of the natural flow regime is altered with significant provided in Annex B through Annex I. negative ecosystem effects. 3.6.3.1 Site Descriptions The Tanzanian system of classification gives a much For each ecological and social component, a thorough smaller range between classes B (somewhat altered) site description was completed highlighting the and C (significantly altered). It does not provide for a unique and important aspects at each site. These site moderate modification with the effect that it might be descriptions provide the foundation for the analysis difficult to classify a system which has undergone some completed in the starter documents, including the site modifications that are beyond the B classification but metrics, indicator selection, management objectives not as bad as class C. It might therefore be necessary and required conditions, which were used in setting to break down the class B into wider classes such as the environmental flow values. Reviewing these site B1, B2, and B3, each representing some increased descriptions also allows individuals who have not level of modification. been to the sites an opportunity to understand the conditions found during the field campaign and the Trajectory of Change (ToC) information used to make the analyses. ToC defines the trend that the system is expected to take assuming no initiative is taken to improve the 3.6.3.2 Site Metrics current/existing condition. It is based on continuation Certain aspects, or metrics, about the system were of the current management practices being carried ranked by each ecological and social component out at a particular site. The trajectories are explained (when possible) to determine the ecological status. The as improving, stable, or declining. method used in classifying the environmental status of the sites has been adapted from the BBM manual Ecological Importance and Sensitivity (EIS) (King, Tharme and Villiers, 2008). The objective EIS expresses the importance of maintaining the of defining the metrics for each site is to categorize diversity of the river and functioning on a local and them based on how much change has happened when national scale and the ability of a river system to resist compared to reference/pristine conditions, how the disturbance or to recover from a disturbance. The change has occurred and what condition these sites present state of a river section does not necessarily are expected to be in the future. The metrics were influences its EIS, however, the level determined for arrived at by considering the following sources of an EIS will have a bearing on what the Ecological information; Management class (EMC) would be for that particular river section. EIS for the sites was is based on three 1. Analysis of data that were collected in the field by classification i.e. the team of specialist 2. Historical/past data that were collected by various - High: Importance of a river in terms of biodiversity organizations in the basin at the national scale 3. Literature review: based on published information - Medium : Importance of a river in terms of on similar studies carried out worldwide biodiversity at the provincial scale 4. Information gathered from the community - Low : Rivers that do not have any uniqueness at any members during field assessment and social studies scale The metrics include: Social Importance and Sensitivity (SIS) Present Ecological State (PES) Defines the social benefits/importance of the system This refers to the present state of the system with to the inhabitants who directly depend on the river relative to the reference condition i.e. how much the system. This considers benefits such as fishing, system has changed when compared to its original water, navigation, hydropower, cultural beliefs etc. state. King et al., 2008 expresses PES in classes The rating is given based on whether the community from A through F, where A represents natural/ utilizes such benefits and not merely by its presence. pristine conditions, B slightly modified, C moderately For example, if a river provides fish but due to various modified, D high degree of modification and E and F reasons such as cultural beliefs and preferences representing a highly modified system that may not prevents the community from eating fish, the SIS
65 of that river may be considered low. Similarly to the aren’t adequate data available. ecological importance classification, SIS is rated as The score for the confidence rankings is given in a high, medium or low. five-point scale: Ecological Management Class (EMC) 1. Marginal to zero confidence. There is almost no This describes the target for ecological management reliable supporting information available. or the desired state for a water resources. The EMC 2. Low confidence: The information available for a resource is usually set last after PES, EIS and indicates some support but may require extensive SIS have already been determined and cannot be research set at a lower class than the PES. The EMC uses the 3. Moderate confidence: Some research may be same classification as PES outlined by King, Tharme necessary and Villiers, 2008 but without the E-F classes which 4. High confidence: Specific issues may need research are considered unsustainable. Similarly to the PES, to confirm the EMC of the sites has been defined based on the 5. Very high confidence: No more information needed. Tanzanian classification system. EMC is linked to the RQOs and reflects the specialist team’s interpretation 3.6.3.6 Data Gaps of the narrative RQOs set by the stakeholders. These are the unanswered questions arising out of the challenges faced when conducting the assessment. 3.6.3.3 Indicators and Management Objectives They represent missing information (which often Indicators refer to the specific parameters that results in a lower confidence ranking) but form the were used to guide the flow recommendations. This basis for future research. These data gaps inform the could be the most sensitive species or species whose environmental flows monitoring plan and adaptive absence or presence would indicate the state of the management so that the proper data can be collected ecosystem, or a critical ecosystem function or habitat to inform future updates to the reserve values condition. For example this can be a type of fish or (and also increase the confidence of the updated macroinvertebrate or plant that is critical to the area, recommendations). or adequate flushing of sediments to provide fish spawning habitat. 3.6.4 Flow Setting Technical Meeting The final activity in the BBM methodology is to These indicators were then linked to management hold a flow setting technical meeting. The objective objectives, or the conditions in which each indicator of this meeting is to set the environmental flow should be maintained. These are important because recommendations for each hydrological component, they synthesize the most important linkages between or building block, at each site. Depending on the flow and the dependent ecology and social uses into various ecological requirements and conditions qualitative statements. These statements can then of each site, the specialists recommend different be compared and combined with the draft RQO habitat requirements related to water (such as depth, statements developed by stakeholders to develop final velocity, and/or inundation) for different hydrological RQO statements to guide management of the natural components or “building blocks” (i.e., low flow driest resources. month, low flow wettest month and high flows, freshets and floods for the maintenance and drought years ). 3.6.3.4 Required Conditions These requirements are then linked to flow using a The required conditions are the physical aquatic hydraulic model and rating curve developed for each habitat requirements necessary to maintain the study site. Each specialist also provides the rationale functioning of the aquatic ecosystem and important for recommending such flows and describes potential social uses. This can be a requirement for flow, depth, consequences of not meeting such flows. Based on velocity, inundation period, etc. and are driven by the recommended flows, the hydrologist then checks the indicators and management objectives. During the recommendations against the hydrological the flow setting technical meeting, these required record and the regionalized data to assess if flows conditions were linked with the hydraulic model to were feasible in terms of occurrences and if it is turn depths and velocities into flow estimates which possible to achieve them (i.e., not unrealistic based were then used to provide the final flow values for the on the hydrological record). From the many flow environmental flow component of the reserve. recommendations given by the individual experts, a consensus must be reached on one recommended 3.6.3.5 Confidence Rankings flow for each building block at each site that meets Confidence rankings are used to identify how much the requirements for all the components. In the end, weight each specialist has attached to the rating given each site will have a monthly low flow value for one for a particular site based on available information hydrological year (for both maintenance and drought and/or expert knowledge. They can be used to identify years) and recommendations for freshets and floods the complexities of a particular site, limitations and/ that together form the final recommendations for the or challenges faced by the specialists in terms of environmental flow. data availability, technique used and timing. These are important to note since river systems are highly The flow setting technical meeting was held from interconnected and complex, and the technical 1st July to 4th July 2019 in Musoma, Tanzania. This experts are often required to make recommendations based on their professional judgement when there
4A maintenance year is considered a “normal” hydrological year where all aspects of ecological function should occur, while a drought year is a very low flow year where species survival is the primary function (King, Tharme and Villiers, 2008)(King, Tharme and Villiers, 2008) 66 meeting included the technical team, partners from the LVBWB, MoW, NBI/NELSAP, and WWF. Meeting n Pf=Pp*(1+i) notes from this meeting can be found in Annex J. where Pf is the future population, Pp is the present 3.7 Reserve Setting and Monitoring population, i is the growth rate, and n is the number 3.7.1 Reserve Values of years of growth. A growth rate for each district in Following the definition in the Water Resources the Lower MRB was provided by the 2012 Census, Management Act of 2009, the reserve includes both with values ranging from 2.2 percent to 3.5 percent the amount of water required for basic human needs and an average value of 2.5 percent (Table 2 3). Since and to protect aquatic ecosystems (environmental the RUs often contain areas of multiple boundaries, flows). the average value of 2.5 percent was used for all calculations. The flow requirements for basic human needs are calculated using the value of 25 liters/person/day As described previously in this section, the values and the population living in the selected planning required for environmental flows were determined area. For this effort, the RUs delineated in the RQO using the NBI E-Flows Framework, with a modified process were used as the planning areas. Population BBM used to determine the flow-ecology-ecosystem data from the 2012 national census were the latest linkages. available with the smallest unit of enumeration being at the ward level (NBS, 2012). To estimate the 3.7.2 Monitoring and Adaptive Management population living in each RU, a spatial layer with Monitoring and adaptive management is a critical the ward population was overlaid with a spatial part of effectively implementing environmental flows. layer containing the boundaries of the RUs and the Monitoring activities should gather information population in each RU was calculated using GIS that answers specific management questions and (geographic information systems). These population also addresses data gaps that have been identified values were then projected to the current planning in the EFA process. The specific activities will need year (2018 for this effort). To estimate growth the to be customized to the management priorities and growth from 2012 to 2018, the following population capacity of the LVBWB, but recommendations have project equation was used:
67 68 4. STUDY SITE RESULTS been provided for monitoring activities and adaptive quite incised with large, vertical, exposed banks management cycles specific for basic human needs and some evidence of bank collapse. The bed level is and environmental flows. controlled by exposed bedrock with areas of riffles, 4.1 Kogatende gravel bars, and sand bars. The vegetation appeared The Mara River at Kogatende site is the most to have recruitment in marginal zone and the lower upstream EFA biophysical study site, which is located sub-zone. The surrounding landscape has a natural within SENAPA close to the Kogatende ranger savannah ecosystem due to its location inside the station and airstrip. It is about 20 km downstream of the Kenya/Tanzanian border. The channel is
FIGURE 4 1: SITE PHOTOS OF KOGATENDE
69 national park, but there are many hippopotamuses 4.1.2 Biophysical – Mara River at Kogatende at the site which may be impacting the water quality 4.1.2.1 Hydrology and embeddedness of the substrate. Kogatende has a contributing upstream catchment 4.1.1 Social Survey area of 8,926 km2 with a dominated land use of Since the Kogatende sub-basin lies almost entirely grass- and shrub land (Figure 4 2). Precipitation within SENAPA and there are no villages present, the social survey was not conducted at this site.
FIGURE 4 2: LAND USE WITHIN KOGATENDE SUB-CATCHMENT (LEFT) AND THE CATCHMENT AREA (RIGHT) in the catchment sums up to 936 millimeters per the driest month is October, although flows can be annum (mm/yr), resulting in an evaporation value of quite low in all months of the year depending on 860 mm/yr, and a runoff value of 76 mm/yr. the rainfall patterns in the upstream portion of the Average monthly, minimum and maximum discharge basin. The graphs are generated from regionalized values for Kogatende EFA site are presented in Figure 4 3. The wettest month is typically May while
FIGURE 4 3: AVERAGE MONTHLY, MINIMUM AND MAXIMUM DISCHARGE VALUES FOR KOGATENDE
70 data as explained in Section 3.3.1. A full analysis of the hydrological assessment can be found in Annex in bedrock. The channel cross sectional profile and C. observed flow levels are given in Figure 4 4. Observed 4.1.2.2 Hydraulics flow velocities against depth show a poor positive The Mara River at Kogatende is a steep bedrock relationship between depth and velocity (R2 =0.1). controlled reach. The bedrock spurs run across Velocity tends to increase with depth, but this is the channel, forming small bedrock steps, chutes, not always the case, indicating that there are areas rapids, runs, short riffles and pools. Faster flows were along transects where deeper water is slow flowing or observed along steeper bedrock sections and riffles, with low flow velocities in pools and smaller hollows
FIGURE 4 4: A SELECTION OF STAGE LEVELS FOR VARIOUS DISCHARGES AT KOGATENDE. OBSERVED DISCHARGES INDICATED BY DOTTED LINE.
FIGURE 4 5: FREQUENCY DISTRIBUTION OF DEPTH-VELOCITY CLASSES FOR THE WETTED CHANNEL FOR THE MARA RIVER AT KOGATENDE
71 very shallow water is fast flowing. This shows large variability of hydraulic habitat along the transect is deposited along the inset benches and on sheltered (Figure 4 5). portions of the bedrock. A thick (5 cm) hippo dung 4.1.2.3 Geomorphology layer is present on the bottom of areas with slow flow The river with a gradient of 0.001727 has a straight and adds a significant covering of organic material channel type with limited sediment storage along over inorganic bed material. the channel margins due to its incised state (Figure A large embedded gravel bar extends from the 4 6). The channel is incised with high steep banks sand bar to the right bank inset bench (Figure 4 7). (Figure 4 7). Active bank erosion is evident along The sand and cobble bar is possibly a tributary bar the right bank and an active flood bench has formed forming downstream of the additional sediment along the left bank. Younger trees, recent fine sand input. The inset bench consists of silt and fine sand deposits and flood debris indicate flood activity at and is sparsely covered with sedges and forbs. The this level. A small inset bench has formed along both right bank is vertical and is actively eroding. banks, consisting of fine sand and silt. The left side This site is fairly natural with almost no modifications of the channel is dominated with bedrock with small caused by human activities or altered flows. The main pockets of mobile gravel in between the bedrock. Silt
FIGURE 4 6: AERIAL VIEW OF KOGATENDE SITE SHOWING THE LOCATION OF VARIOUS GEOMORPHIC FEATURES AND THE LOCATION OF THE CROSS SECTION.
72 FIGURE 4 7: CROSS SECTION OF THE KOGATENDE SITE SHOWING THE GEOMORPHIC FEATURES, SEDIMENT DEPOSITION AND VEGETATION TYPES. ecological factor driving vegetation in addition to Bedrock outcrops form refugia for forb and shrub flows is grazing by mega herbivores, such as hippos. species. Figure 4 8 shows the distribution of plant species along the cross section. 4.1.2.4 Riparian Vegetation 4.1.2.5 Fish The left bank was more vegetated than right bank The Mara River at Kogatende is relatively straight with many forbs, grasses and small Acacia trees. with water flowing in a north-westerly direction. Two At the lower parts there were few plants including primary GHUs were identified and delineated for the a mixture of grasses, sedges and forbs. These were site, namely a pool and a run. A total of five pools, such as Eleusine coracana, Echinochloa haploclada comprising three shallow and two deep pools were (grasses), Cyperus distans (sedge), and Commelina sampled with an electro-fisher, all with a slow velocity benghalensis and Sphaeranthus steetzii (forbs). The (Figure 4 9). The two velocity depths classes include upper part was covered largely with forb species SD and SS. Two runs were identified and delineated, including Harpachne schimperi, Schkuhria pinnata with one run being assigned a velocity depth class of and Parthenium hysterophorus - an invasive weed fast shallow and the other run classes as slow shallow. and trees like Croton dichogamous and Acacia spp. The substrate is dominated by bedrock and boulders, There was a flood bench (left side) which was covered with sand and gravel also present across the reach. with mainly grasses of Cynodon nlemfuensis and No aquatic or marginal vegetation was present at Dactyloctenium aegyptium species. The site was the site, with overhanging vegetation and roots actively used by hippos and crocodiles at the time absent from the reach. There was also no evidence of of field survey including using pools. Banks were undercut banks. A thick (5 cm) hippo dung layer was covered with vegetation by almost 40 percent. present on the bottom of areas with slow flow, notably the pool areas. Dung was also present in the water
73 FIGURE 4 8: CROSS SECTION OF THE MARA RIVER AT KOGATENDE SHOWING INDICATOR PLANT SPECIES
TABLE 4 1: A SUMMARY OF THE HCR FOR THE REACH WITH ASSOCIATED VELOCITY DEPTH CLASS RATINGS AND CORRESPONDING DETAILS
74 FIGURE 4 9: AERIAL VIEW OF KOGATENDE SHOWING THE DELINEATED GHUS FOR THE SITE (GOOGLE EARTH)
FIGURE 4 10: A PHOTOGRAPH DEPICTING HABITAT CHARACTERISTICS FOR THE KOGATENDE REACH (FEBRUARY, 2019) 75 column, with the eutrophic levels expected to be Habitat diversity was considered to be relatively high, relatively high. An overview of the HCR ratings is with variable velocity-depth classes, but with limited presented in Table 4 1 and Figure 4 10. substrate and habitat cover available. A total of nine fish species were sampled from the site, with a total of 62 individuals recorded for the 4.1.2.6 Macroinvertebrates site. One indicator species were recorded for the Most of the macroinvertebrate taxa are moderately to site, namely Oreochromis niloticus. The species highly sensitive to river flow and habitat availability composition was dominated by Labeo sp and (i.e. Naucoridae, Gomphidae, Lestidae, Baetidae, Enteromius sp. Caenidae, Simuliidae, Elmidae, Tricorythidae and Substrate was dominated by bedrock, with Hydropsychidae), and some are very sensitive to poor rocks, cobbles and gravel also recorded for the water quality (Oligoneuridae). Table 4 2 and Table 4 site. Habitat cover is predominantly associated 3 below indicate the field processed and laboratory with the identified substrate and water column. processed score results for the site. Despite the Bedrock was covered in silt and sand which has presence of some sensitive taxa such as Oligoneuridae, a negative impact on the diatom periphyton layer. these were in low abundances and there has been
TABLE 4 2: MACROINVERTEBRATE COMMUNITY METRICS FOR KOGATENDE
TABLE 4 3: RESULTS OF THE SASS5 AND TARISS ON PROCESSED SAMPLES AT KOGATENDE
76 evidence of these disappearing and being replaced by high conductivity. The high conductivity could also tolerant taxa such as Diptera and Oligochaeta during be a factor of hippo dung which would also influence periods of extreme low flows. the turbidity of the water. 4.1.2.7 Water Quality Turbidity and conductivity were quite high at this According to Subalusky et al., 2018, organic matter site but the temperature and pH were within the from hippos increase the nutrient loading in rivers limits set by the Tanzania Bureau of Standards for which could result in low oxygen. As a result of the drinking water. Conductivity increases with increase hippo influence, highest level of organic matter was in temperature. This is because with increase in recorded at this site in comparison to the other sites temperature, more water evaporates from the surface during the 1st survey. However, no detectable levels of the river leaving behind salts which contributes to of nitrate was measured. This could have been due to nitrate denitrification resulting from the low
TABLE 4 4: RESULTS OF WATER QUALITY ANALYSIS AT KOGATENDE oxygen levels in the pools. Table 4 4 shows the field
TABLE 4 5: SITE METRICS FOR KOGATENDE
77 measurements and laboratory results for water
TABLE 4 6: INDICATORS AND MANAGEMENT OBJECTIVES FOR KOGATENDE
78 quality samples taken at the site. Table 4 7: Required flow conditions for Kogatende
79 4.1.3 Site Metrics 4.1.4 Indicators and Management Objectives TABLE 4 8: CONFIDENCE LEVEL FOR FLOW RECOMMENDATIONS FOR KOGATENDE
80 4.1.5 Required Conditions 4.1.6 Confidence Ratings rise and fall quickly. There are fine and course sand on the bed along with larger cobles and gravels. There 4.2 Tobora are good refuge areas for aquatic species, like fish and The Tobora biophysical site is located at the macroinvertebrates. No resident fish species found, downstream end of the Tobora River, about 2.5 km indicating this river is only used temporarily by fish upstream from where it joins the mainstem Mara species. There are likely springs and groundwater River. It is a straight channel that is confined to flow that maintain at least a small flow in the river the valley bottom due to geology of the area, with during a normal year. Agricultural fields go right up evidence of slight incision. The system is likely very to the banks and there is active sand mining in this flashy, with rainfall events causing the water level to river.
FIGURE 4 11: SITE PHOTOS OF TOBORA 81 species identified), tree species that provide fruits (6 species identified), timber and poles (22 species), 4.2.1 Social Survey – Matare, Nyamerama weaving and thatching materials. The river condition Tobora River is somewhat degraded, but it has some is relatively good but the trend suggests that due to natural conditions necessary for ecosystem goods and increasing population and anthropological activities services enhancement. The river Inundations occur including deforestation and cultivation along the mainly at the downstream and more specifically to its river banks, degradation of the river will accelerate tributaries; which last at most three days, extending in the future. to a width of about one to three metres mainly in April when the rain is at its maximum. The diversity Among the resources available in the RU, water was of fish species is much less compared to other sites. reported to be the most important accounting for an It was mentioned that only two species exist; Mumi average about 74 percent with significant importance and Ningu, mostly available during the flooding recorded in Nyamerama village (98.7 percent) as season of April and May. The area has fairly diverse opposed to Matare (49.7 percent). Crop cultivation riparian vegetation that is important for locals’ was rated second (12.4 percent), followed by livestock livelihoods. There is a lot of natural vegetables (11
TABLE 4 9: WETLAND RESOURCES AND THEIR RELATIVE IMPORTANCE TO THE LIVELIHOOD OF COMMUNITIES IN TOBORA
82 pastures (6.7 percent), building poles (3 percent) and finger millet, cassava and sweet potatoes. Petty natural vegetables (1.6 percent). Other resources trade involves running small-scale retail businesses accounted for less than 1 percent (Table 4 9). dealing in household stuff. Economic activities are classified under crop Assessing how the trend of resources utilization and farming, livestock keeping and petty trade (Table condition of the wetland resources have changed 4 10). Farming is common in both villages and the over the years was done and the results indicated in major crops cultivated include maize, sorghum, Table 4 11 below. A hypothetical value of 100 was set as a current benchmark and the respondents asked
TABLE 4 10: ECONOMIC ACTIVITIES IN TOBORA RU
TABLE 4 11: TREND OF RESOURCES AND CONDITIONS OF THE WETLAND IN TOBORA
83 to give their perceived values in the past and future, with reference to the set benchmark. The possible reasons for change were also highlighted. 4.2.2. Biophysical – Tobora River at catchment area of 361 km2 with a dominated land Nyasurura use of agriculture, savanna, and grassland (Figure 4.2.2.1 Hydrology The EFA site at Tobora has a contributing upstream
FIGURE 4 12: LAND USE WITHIN THE TOBORA SUB-CATCHMENT (LEFT) AND CATCHMENT AREA (RIGHT)
4 12). Precipitation in the catchment sums up to 945 values for Tobora EFA site are presented in Figure mm/yr, resulting in an evaporation value of 869 mm/ 4 13. On average, the wettest month is May while yr, and a runoff value of 77 mm/yr. the driest month is August. The flow in the Tobora Average monthly, minimum and maximum discharge River can reduce to zero, particularly in drought
FIGURE 4 13: AVERAGE MONTHLY, MINIMUM AND MAXIMUM DISCHARGE VALUES FOR TOBORA 84 years, where water remains only in pools that are refreshed by groundwater or sub-surface flow. turbulent flows. Good vegetation and root structure leads to slower flows along the banks of pools and 4.2.2.2 Hydraulics riffle sections. The shape of the cross section and The Tobora River reach upstream of the confluence observed and modelled flow levels are given in Figure with the Mara River is a steep bedrock controlled 4 14. The relationship between velocity and depth channel with small pools/areas with slower flow was weak for the low flow (R2 = 0.28) and higher flow amongst the faster riffles and runs. Cobbles and gravel (R2 = 0.33). Velocity tends to increase with depth, are present along the steeper riffle section creating but this is not always the case, showing that there faster turbulent flows. The flatter pool sections have are areas along transects where deeper water is slow a sandy or silty bed resulting in deeper and less flowing or very shallow water is fast flowing. This
FIGURE 4 14: CROSS SECTION SHOWING OBSERVED AND MODELLED FLOW DATA FOR TOBORA
FIGURE 4 15: FREQUENCY DISTRIBUTION OF VELOCITY-DEPTH CLASSES FOR TOBORA 85 shows some of the variability of hydraulic habitat along the transect. The modelled velocity-depth class site. The river appears to be incised with terraces frequency distributions are given in Figure 4 15. along both banks and narrow flood benches (Figure 4.2.2.3 Geomorphology 4 17). A steady sand supply is evident from the thick The Tobora River site has a pool-rapid sequence with sand deposits on the flood benches. Small inset pools and riffles of roughly equal length (Figure benches form along the margins, with dense sedge 4 16). The reach is bedrock controlled and has a growth and provide good marginal cover. Sand is the local gradient of 0.0057 and can be classified as the dominant sediment type, with the banks consisting upper foothill zone (Rowntree and Wadeson, 1999). of fine sand, the benches of medium sand and the bed Rowntree and Wadeson describe the reference of coarse sand and fine gravel along pool sections and condition for this gradient as ‘moderately steep, large gravel and cobble along riffle sections. Cobble cobble bed or mixed bedrock-cobble bed channel, and gravel voids are filled with coarse sand along the with plain bed, pool-riffle or pool-rapid reach types. riffle. The length of pools and riffles are similar, and a narrow floodplain of sand, gravel or cobble often The river level was high and turbid with sand actively present’. The site fits this description well. moving on the bed during the sampling trip in February 2019. Cultivation extends down to the river The site is located 200 m upstream of a bridge, with banks and sand mining is taking place along sand no clear effect on the depositional features at the
FIGURE 4 16: AERIAL VIEW OF THE TOBORA RIVER SHOWING THE LOCATION OF THE CROSS SECTION, POOLS AND RIFFLES (GOOGLE EARTH IMAGE 24 SEPTEMBER 2010).
86 FIGURE 4 17: A CROSS SECTIONAL VIEW OF THE TOBORA RIVER INDICATING GEOMORPHOLOGICAL FEATURES, SEDIMENT COMPOSITION AND VEGETATION TYPES
bars and benches. Sand is harvested with buckets with herbaceous vegetation, trees (e.g., Ficus sur from sand bars without much impact on the instream and Grewia similis) were observed on both banks habitat. which provide support to stabilize banks. Marginal 4.2.2.4 Riparian Vegetation sub-zone was dominated by papyrus plants Cyperus Both the left and right banks of the Tobora River were involutus and Cyperus distans while lower part was vegetated with different plant species. Reference occupied by species like Dactyloctenium aegyptium, condition of this site would be the same as PES but with Panicum maximum and Eragrostis ciliaris. The left less influence from human activities. Consequently, bank had more shrub species dominated by Lantana less or no invasive and exotic species on lower and camara which appear to be grazed by goats and sheep, upper sub-zones was to be expected. The banks and and Tithonia diversifolia. Cynodon nlemfuensis was macro-channel bank should be more covered with present on the sand bars on the left side of the river. vegetation and especially trees, as opposed to the At this site devil’s plant which is an invasive species current situation. Several human activities were was also observed on the MACRO-CHANNEL BANK. observed during field surveys including sand mining Banks were covered by vegetation by 70 percent. Sand and crop farming/vegetable gardening close to the which likely was coming from the river and farms river about 6 m. In addition to banks being covered was deposited at the upper right bank. Distribution
87 FIGURE 4 18: CROSS SECTION OF THE TOBORA RIVER SHOWING INDICATOR PLANT SPECIES
of plant species along the cross section is indicated in Figure 4 18. the substrate associated with the riffles and runs characterized by rocks, cobbles and boulders. 4.2.2.5 Fish Turbidity was considered to be high. Limited aquatic Tobora River meanders with the flow in a westerly, vegetation was present at the site, with marginal and north-westerly direction. Three primary GHUs were overhanging vegetation present for almost the entire identified and delineated for the site, namely a pool, riffle and a run. A total of four pools, two riffles and two runs were sampled with an electro-fisher, all with a shallow depth (Figure 4 19). The velocity depth classes associated with the pool was SS, with the velocity depth classes associated with the riffle and run GHUs all being classed as FS. Substrate in the pools is dominated by sand and gravel, with
88 FIGURE 4 19: AERIAL VIEW OF THE TOBORA RIVER SHOWING THE DELINEATED GHUS FOR THE SITE (GOOGLE EARTH)
TABLE 4 12: A SUMMARY OF THE HCR FOR THE REACH WITH ASSOCIATED VELOCITY DEPTH CLASS RATINGS AND CORRESPONDING DETAILS.
89 FIGURE 4 20: A PHOTOGRAPH DEPICTING HABITAT CHARACTERISTICS FOR THE TOBORA RIVER REACH (FEBRUARY 2019) extent of the reach. Root wads and undercut banks activity and watering, some species of Baetidae and were also recorded. An overview of the HCR ratings Heptageniidae were among the sensitive taxa recorded is presented in Table 4 12 and Figure 4 20. at the site. Signs of flow modifications include low A total of four fish species were sampled from the site, abundances of flow sensitive taxa (Tricorythidae, with a total of 22 individuals recorded for the site. Simuliidae and Hydropsychidae) which form 27 No indicator species were recorded for the site. The percent of all individuals at the site. The site is species composition was dominated by Labeo sp and impacted by sand harvesting and cultivation of the Enteromius sp. riparian zone. Small –scale irrigation is also done at the site and along the river. Table 4 13 and Table 4 Substrate was variable and consisted of sand, gravel, cobbles, boulders and bedrock. Habitat cover was dominated by undercut banks and overhanging vegetation, with marginal vegetation and root wads also present. Habitat diversity was considered to be relatively high, with two dominant velocity-depth classes, namely SS and FS. Excessive sediment (gravel and sand) is present in the channel. 4.2.2.6 Macroinvertebrates Although there are clear signs of flow alteration in the system, likely linked to land use change and livestock
90 14 below indicate the field processed and laboratory processed score results for the site. acceptable limits. This was also the case for electrical conductivity with low values in the 1st survey due to 4.2.2.7 Water Quality dilution effect and high values on the 2nd survey. Turbidity was the highest at this site during the 1st Generally, the pH at this site is slightly higher than any survey attributed to increased sediments as a result other site probably influenced by geological factors of the rains that had been received the previous day. such as the rock/soil type in the area. Contribution The levels were much higher than the recommended of groundwater during the 2nd survey could be the Tanzanian drinking water standards of 25 NTU. High reason for the high pH compared to the 1st survey turbidity reduce the amount of light that can penetrate when the pH of the river was influenced by pH of the water surface which affects photosynthesis of the rain recorded the previous day. Groundwater aquatic plants therefore reduces the amount of normally has more contact time with the bedrock dissolved oxygen in the water. It also leads to loss of hence affecting the quality of the water based on vision for fish hence affecting its ability to catch prey the bedrock characteristics. Therefore during base and clogging of fish gills resulting in death. However, flow conditions, the pH of the river will most likely due to the flowing water, dissolved oxygen was within reflect that of groundwater than periods of increased
91 rainfall. Table 4 15 shows the field measurements and
92 93 94 laboratory results for water quality samples taken at
95 the site.
4.2.3 Site Metrics 4.2.5 Required Conditions Mara River. It has a mostly rocky substrate with areas 4.2.6 Confidence Ratings of cobbles and sand. The Somoche Sub-basin is larger 4.3 Somoche than the Tobora Sub-Basin but it is drier, resulting The EFA biophysical field site at Somoche is located in higher discharges but a less flashy system. More at the downstream end of the Somoche River, about fish species were found, indicating that this tributary 0.35 km upstream from where it joins the mainstem is important for maintaining biodiversity within
96 grasses at 6 percent, livestock pastures and firewood at 3.7 percent and the rest accounting for less than 2 percent.
The three main economic activities in the villages include agriculture involving about 60 percent of the population, following by livestock keeping at 20 percent and a small percentage of the people run small businesses. Table 4 21 below indicates the status of these activities. Similarly to Tobora RU, the trend of resource utilization and the condition of the wetland in the past (1960 to 1985) to the present and future time (2030) was assessed and responses given in Table 4 22.
Biophysical – Somoche River at Somoche Hydrology The EFA site Somoche has a contributing upstream catchment area of 690 km2 with a dominated land use of agriculture, savanna, and grassland (Figure 4 22). Precipitation in the catchment sums up to 957
FIGURE 4 21: SITE PHOTOS OF SOMOCHE the mainstem Mara River. The surrounding area is dominated by rainfed agriculture and barren areas. Ebikoro) when compared to Tobora which had only two. There are natural vegetables (nine species), tree 4.3.1 Social Survey – Kwitete, Nyamitita species which are important for fruits (17 species), Somoche River does not flow throughout the year timber and poles (39 species). The locals depend but some of its tributaries are perennial. The river on the river for crop farming as well as livestock floods twice a year in April and December, during keeping which are the key livelihood activities in the the wet seasons and can last from one to seven site. Table 4 20 summarizes the wetland resources days extending up to a width of about 100 meters. that are important to communities in Somoche. Somoche River was identified to have more fish Water is the most important resource accounting for species (Mumi, Kamongo, Gogogo, Perege and 77.7 percent followed by building poles and roofing
TABLE 4 20: WETLAND RESOURCES AND THEIR RELATIVE IMPORTANCE TO THE LIVELIHOOD OF COMMUNITIES IN SOMOCHE RU 97 mm/yr, resulting in an evaporation value of 909 mm/ percent and a small percentage of the people run yr, and a runoff value of 48 mm/yr. small businesses. Table 4 21 below indicates the status of these activities. The three main economic activities in the villages include agriculture involving about 60 percent of Similarly to Tobora RU, the trend of resource the population, following by livestock keeping at 20 utilization and the condition of the wetland in the
98 99 past (1960 to 1985) to the present and future time (2030) was assessed and responses given in Table 4 22 4.3.2 Biophysical – Somoche River at The EFA site Somoche has a contributing upstream Somoche catchment area of 690 km2 with a dominated land 4.3.2.1 Hydrology use of agriculture, savanna, and grassland (Figure 4
Figure 4 22: Land use within the Somoche sub-catchment (left) and catchment area (right)
22). Precipitation in the catchment sums up to 957 mm/yr, resulting in an evaporation value of 909 mm/yr, and a runoff value of 48 mm/yr. Average monthly, minimum and maximum discharge values for Somoche EFA site are presented in Figure
FIGURE 4 23: AVERAGE MONTHLY, MINIMUM AND MAXIMUM DISCHARGE VALUES FOR SOMOCHE
100 4 23. On average, the wettest month is May and the driest month is August, regularly going to almost of 0,011 and can be classified as the upper foothill zero during the dry season. zone (Rowntree and Wadeson, 1999). Rowntree and 4.3.2.2 Hydraulics Wadeson describe the reference condition for this The lower Somoche River is a steep fast-flowing gradient class as ‘moderately steep, cobble bed or bedrock channel with slower flowing deep pools. mixed bedrock-cobble bed channel, with plain bed, Large cobbles are present on the bed with small pool-riffle or pool-rapid reach types. The length of pockets where gravel is deposited. Sand and finer pools and riffles are similar, and a narrow floodplain material are deposited on the well vegetated inset of sand, gravel or cobble often present’. The site fits benches. Turbulent fast flowing rapids, riffles the reference description moderately due to the high and runs and slower flowing less turbulent pools terrace and strong bedrock influence. alternate along this reach. The channel cross section shows the observed and modelled flow levels (Figure The channel is incised with a high terrace along the 4 24). Observed velocity-depth data show a very right bank and consists of silt (Figure 4 27). The weak relationship (R2 = <0.1), indicating a partial left bank is sloping and has no clear fluvial features agreement between depth and expected velocity. (consists of fine sand). A vegetated core bar forms a Velocity tends to increase with depth, but the large island along the left side of the channel. Small variability is very high, showing that there are areas medium sand inset benches form along the margins along transects where deeper water is slow flowing and are not vegetated. The channel bed consists of or very shallow water is fast flowing. This shows high bedrock, blocky boulders and slabby cobbles and variability of hydraulic habitat along the transect. gravels along riffles and rapids (Figure 4 27). Pools The frequency distribution for the various velocity- are long with overhanging vegetation. Sand deposits depth classes is shown in Figure 4 25. (medium and coarse sand) form on cobble islands with dense sedges and forbs. Very little evidence 4.3.2.3 Geomorphology of high silt and clay loads on flood features such The Somoche EFA site is located along a straight as sand bars. Limited bank erosion was observed reach with a pool-riffle sequence and strong bedrock despite grazing that is taking place along the banks. influence (Figure 4 26). The site has a local gradient Agriculture takes place up to a meter of the right
FIGURE 4 24: CROSS SECTION SHOWING OBSERVED (DOTTED LINES) AND MODELLED (SOLID LINES) WATER LEVELS FOR SOMOCHE
101 FIGURE 4 25: FREQUENCY DISTRIBUTION OF VELOCITY-DEPTH CLASSES FOR THE WETTED SOMOCHE RIVER CHANNE
FIGURE 4 26: A SATELLITE IMAGE OF THE SOMOCHE RIVER CHANNEL INDICATING THE LOCATION OF TRANSECT ACROSS A RIFFLE (GOOGLE EARTH IMAGE 21 JULY 2017).
102 FIGURE 4 27: A CROSS SECTION ACROSS THE SOMOCHE RIVER SHOWING GEOMORPHIC FEATURES, DOMINANT SEDIMENT COMPOSITION AND VEGETATION TYPES. bank, not allowing much lateral buffering. river is used for grazing while the right is used for crop farming. Riparian indigenous species were 4.3.2.4 Riparian Vegetation found to be abundant i.e. around 65 percent of all The Somoche site was surveyed on both sides of the the species. The site was used as watering point banks and the riparian zone was well covered with for cattle during the field survey. The left bank was vegetation. Ideally the site would have a wider strip more covered with vegetation as compared to right of riparian vegetation with more cover of grasses bank. Overall, the riparian zone was narrow likely and sedges at the marginal and lower sub-zones and due small size of the river itself and human pressure trees on upper sub-zones, at its reference conditions. from grazing and agriculture. Marginal sub-zone Invasive and exotic species expected were expected had species like Paspalum scrobiculatum in addition to be absent at its reference condition. The river flows to Cyperus species. Lower subzone had grass species over a bedrock and large stones and there are also Brachiaria scalaris, Echinochloa pyramidalis and “island” made up of sand bars with several plant Commelina benghalensis. At the upper sub-zone fig species dominated by notably papyrus Cyperus trees Ficus sur and F. exasperata and Acacia sp. were involutus and Cyperus distans. The left side of the present. Sesbania macrantha and Mimosa pigra (an
103 FIGURE 4 28: CROSS SECTION OF THE SOMOCHE RIVER SHOWING INDICATOR PLANT SPECIES
invasive species) occupied benches of the Somoche River. Figure 4 28 shows the cross section with electro-fisher. The velocity depth class associated indicator species. with the GHU was FS. Substrate is dominated by 4.3.2.5 Fish gravel and cobbles, with boulders and bedrock also The Somoche site is located in the tributary of the intermittently present across the reach. Aquatic Mara River, with northerly flow direction. The reach vegetation was present at the site, with marginal is considered to be straight. One primary GHU and overhanging vegetation present. Root wads was identified and delineated for the site, namely and undercut banks were also recorded but limited. a riffle with limited rapid characteristics (Figure 4 29). Sampling was undertaken by means of an
104 FIGURE 4 29: AERIAL VIEW OF THE SOMOCHE REACH SHOWING THE DELINEATED GHU FOR THE SITE (GOOGLE EARTH)
TABLE 4 23: A SUMMARY OF THE HCR FOR THE REACH WITH ASSOCIATED VELOCITY DEPTH CLASS RATINGS AND CORRESPONDING DETAILS.
Turbidity was considered to be high. An overview of variabilis, both classified as critically endangered. the HCR ratings is presented in Table 4 23 and Figure The species composition was dominated by Labeo sp, 4 30. comprising three genus. A total of 14 fish species were sampled from the Substrate was variable and consisted of numerous site, with a total of 108 individuals recorded for the substrate types, with gravel and cobbles being site. Two indicator species were recorded for the dominant. Habitat cover was dominated by undercut site, namely Labeo victorianus and Oreochromis
105 FIGURE 4 30: A PHOTOGRAPH DEPICTING HABITAT CHARACTERISTICS FOR THE SOMOCHE REACH (FEBRUARY 2019)
banks and also aquatic and overhanging vegetation. Habitat diversity was considered to be high, with one activity and drinking, > 3 species of baetidae and dominant velocity-depth class, namely FS. Heptageniidae and Oligoneuridae were recorded 4.3.2.6 Macroinvertebrates at the site. This is also the only site that recorded Although there are clear signs of flow alteration in the freshwater crabs (Potamonautidae). The site recorded system, likely linked to land use change and livestock six rheophilic families which form 51 percent of all
106 individuals at the site. Table 4 24 and Table 4 25 below indicate the field and laboratory processed Turbidity was much higher during the 2nd round of score results for the site. survey than the 1st due to increased rainfall therefore 4.3.2.7 Water Quality increased sediments in the river. Small levels of EC, pH, DO of the site within acceptable limits for total nitrogen measured but also within acceptable both drinking water standards and fish requirements.
107 limits. Table 4 26 shows the field measurements and laboratory results for water quality samples taken at
the site. 4.3.3 Site Metrics
108 109 4.3.4 Indicators and Management Objectives
110 111 4.3.5 Required Conditions although it is still considered a seasonal system by 4.3.6 Confidence Ratings local residents. The Tigithe River is dominated by a 4.4 Tigithe riffle-pool system, with fine gravels, silty clays, and The EFA biophysical site at Tigithe is located about 5 large cobbles present. There are high terraces on km upstream from where it flows into the upper Mara both banks and slight evidence of incision. There is Wetland. It flows parallel to the Mara River and flows commercial and artisanal mining in this catchment directly into the Mara Wetland, near to where the and there is concern over pollution in the river, Mara River enters the wetland. The area has higher although this is not related to alterations in flow. The rainfall than other areas of the Mara River Basin, fish here seem highly connected with the wetland species, indicating migration between the two
FIGURE 4 31: SITE PHOTOS OF TIGITHE
112 areas. Flows during drought conditions are needed wetland which its extension has been increasing over to maintain good water quality in pools for aquatic the years. species. 4.4.1 Social Survey 4.4.1.1 Lower Tigithe The Tigithe sub-basin consists of two RUs: the Lower Lower Tigithe has about 10 species of fish that were Tigithe RU and the Upper Tigithe RU. Tigithe River identified by the community, 17 species of natural is seasonal and dries up three times yearly from July vegetables, 17 species of fruit trees, 50 tree species to August. However, it is fed by many tributaries for building poles, five species of weaving reeds and which are either permanent or seasonal. Flooding seven species of thatching grass. Fourteen ecosystem in the Lower Tigithe RU mostly occurs in the lower services derived from wetland and rivers were and middle reaches of the rivers due to confluence identified to be useful for livelihood of communities of various tributaries. The inundation lasts for a day in Lower Tigithe RU. Table 4 31 shows the wetland and extends up to three kilometers while the depth of water goes up to one meter. This has formed a
resources/ecosystem services obtained in this RU Matongo village (Table 4 32). Agricultural activities and their relative importance to the two villages. involve cultivation of crops such as maize, millet, bullrush millet, cassava, sweet potatoes, vegetables, Four main economic activities are practiced in this RU i.e. Agriculture, livestock keeping (cows, sheep, chicken and donkey), petty trade, and mining in
113 114 paddy/rice and bananas. The trend of resource utilization and condition of the wetland from the past in Lower Tigithe. This include two fish species (Mumi to the future is shown in Table 4 33 below. and Furu), 12 species of natural vegetables, 12 species 4.4.1.2 Upper Tigithe of fruit trees, 10 species of timber and building poles, The number of resources reported to be available in one weaving material and four types of thatching Upper Tigithe is much less compared to the number
115 grass. The number of wetland resources and their importance to the community is outlined in Table 4 34 below. In terms of economic activities, only three were
identified in the village: agriculture, livestock keeping and condition of the wetland resources over the and petty trade (Table 4 35). years. A hypothetical value of 100 was set as a current benchmark and perceived values of these Table 4 36 below shows trend of resources utilization
116 resources in the past and future with reference to the set benchmark indicated. The possible reasons for change have also highlighted. The Tigithe EFA site has a contributing upstream 4.4.2 Biophysical – Tigithe River at Matongo catchment area of 183 km2 with a dominated land 4.4.2.1 Hydrology use of agriculture, shrub land, and savanna (Figure 4
FIGURE 4 32: LAND USE WITHIN THE TIGITHE SUB-CATCHMENT (LEFT) AND CATCHMENT AREA (RIGHT)
117 32). Precipitation in the catchment sums up to 1,100 mm/yr, resulting in an evaporation value of 1,018 mm/yr, and a runoff value of 82 mm/yr. Average monthly, minimum and maximum discharge
FIGURE 4 33: AVERAGE MONTHLY, MINIMUM AND MAXIMUM DISCHARGE VALUES FOR THE TIGITHE EFA SITE values for Tigithe EFA site are presented in Figure 4 33. Typically, April is the wettest month and August in Figure 4 34 with the observed and modelled flow is the driest month, going almost to zero m3/s. levels indicated. There was a weak relationship 4.4.2.2 Hydraulics between depth and velocity (R2 < 0.36), showing The Tigithe River follows a pool riffle sequence with high variability in the velocity–depth association slow-flowing pools and steeper faster turbulent flow along riffle sections. The hydraulic transect is shown
FIGURE 4 34: CROSS SECTION FOR TIGITHE RIVER SHOWING OBSERVED FLOW LEVELS (DOTTED LINES) AND MODELLED FLOW LEVELS (SOLID LINES)
118 FIGURE 4 35: FREQUENCY DISTRIBUTION FOR VELOCITY-DEPTH CLASSES FOR THE WETTED CHANNEL OF THE TIGITHE RIVER across the channel. The frequency distribution of the flow velocity-depth classes for the wetted channel are consists of silt and the floodplain/bench of layers of presented in Figure 4 35. small gravel and sand. Recent flood deposited sand 4.4.2.3 Geomorphology is present on the floodplain/bench. The banks are Tigithe channel is a straight to sinuous channel with near vertical with active erosion along short sections a pool-rapid sequence (Figure 4 36). The channel of bank. A narrow inset bench lines the right bank slope is 0.007808 and can be classified as a channel composed of medium sand. The riffles consist of of the upper foothill zone (Rowntree and Wadeson, armored cobble with voids filled with coarse sand 1999). Rowntree and Wadeson define the reference and fine gravel. A silt drape is present on bed features condition as “moderately steep, cobble bed or mixed where the flow velocities are lower. The bed of the bedrock-cobble bed channel with plain bed, pool pools consist of fine gravel and silt. riffle, or pool rapid reach types. The length of pools and riffles /rapids are similar. Narrow flood plain of The Tigithe River plays an important role in sand, gravel or cobble often present”. The site fits this maintaining a large back swamp area to the North description well. of the Mara channel (Figure 4 38). The back swamp forms on the Mara floodplain as a result of the alluvial The channel is bedrock controlled and incised into the ridge that forms due to sediment deposition along landscape as is evident by the narrow floodplain and the Mara River. The Mara spills into this back swamp terrace along the right bank (Figure 4 37). The terrace during flood conditions, but the Tigithe permanently
119 FIGURE 4 36: A SATELLITE IMAGE SHOWING THE TRANSECT POSITION (WHITE LINE) AND THE POOL-RIFFLE SEQUENCE ALONG THE TIGITHE RIVER (GOOGLE EARTH IMAGE DATE 17 JAN 2010).
Figure 4 37: Channel cross section for the Tigithe River indicating the geomorphic features (black text), sediment type (brown text) and vegetation type (green text). 120 FIGURE 4 38: A GOOGLE EARTH IMAGE SHOWING THE BACK SWAMP (ENCIRCLED BY THE WHITE LINE) FORMED BY THE ALLUVIAL RIDGE/LEVEE (BROWN LINE) ALONG MARA RIVER. THE TIGITHE RIVER DRAINS INTO THE BACK SWAMP PERMANENTLY AND THE MARA RIVER ONLY DURING FLOOD FLOWS. FLOW DIRECTION IS FROM EAST TO WEST. contributes to its water balance, making it a crucial source of surface water during low flow conditions. brizantha. Grazing and browsing by livestock appear to be high resulting in to lawns on the floodplain 4.4.2.4 Riparian Vegetation and pruned shrubs on terrestrial part of the system. The bed of Tigithe River is made up of cobbles stones Lantana camara was present in macro-channel bank. and sand and both left and right banks appear to be Overall, the vegetation cover was about 70 percent. stable. At its natural state the site would have more Figure 4 39 shows cross section and plant species vegetation cover in riparian area and at both banks. distribution at the site. Herbaceous vegetation would have been tall on the floodplain which is on left side of the river. There 4.4.2.5 Fish would also be more riparian trees on both river banks. Tigithe River meanders with the flow in a westerly Absence of terrestrial tree species like Leucacena and northerly direction for the reach surveyed. Two leucocephala and Lantana camara on the macro- primary GHUs were identified and delineated for channel bank would be expected. A strip of around the site, namely a pool and rapids (Figure 4 40). A 100 m was selected to study riparian vegetation in total of four shallow pools and one deep pool were Tigithe. sampled with an electro-fisher and fyke net for the study, with all pools characterized by slow velocity. Members of Poacea and Cyperaceae families Four rapid areas were sampled, with all these areas dominated the marginal sub-zone including Cyperus characterized by a FS velocity-depth class. Substrate cyperoides, Coix lacryma and Eriochloa macclounii. in the pools is dominated by sand and cobbles, with Lower sub-zone was occupied by Commelina the substrate associated with the rapids characterized carsonii, C. benghalensis, Cyperus cyperoides and by rocks, cobbles and bedrock. Cynodon dactylon. Among the studied sites this was the site with more species richness in the Mara Turbidity was considered to be high. No aquatic basin. Floodplain area was present on the left side vegetation was present at the site, with marginal and and was dominated by grasses including Cynodon overhanging vegetation present for almost the entire nlemfuensis (around 60 percent), Setaria sphacelata, extent of the reach. Root wads and undercut banks Echinochloa sp., Hyparrhenia rufa and Brachiaria were also recorded for the majority of the pool units.
121 FIGURE 4 39: CROSS SECTION OF THE TIGITHE RIVER SHOWING INDICATOR PLANT SPECIES
122 FIGURE 4 40: AERIAL VIEW OF THE TIGITHE RIVER SHOWING THE DELINEATED GHUS FOR THE SITE (GOOGLE EARTH)
FIGURE 4 41: A PHOTOGRAPH DEPICTING HABITAT CHARACTERISTICS FOR THE TIGITHE RIVER REACH (FEBRUARY 2019)
123 Table 4 37 and Figure 4 41 show an overview of the Some of sensitive macroinvertebrate taxa were HCR ratings. collected at this site (i.e. > 2 species of Baetidae, Tricorythidae and Hydropsychidae), and some that are A total of 15 fish species were sampled from the site, very sensitive to poor water quality (Heptageniidae, with a total of 124 individuals recorded for the site. Leptophlebiidae). The site is affected by mining and Two indicator species were recorded for the site, discharge of wastewater from mining, erosion and namely Labeo victorianus and Clarias gariepinus, sedimentation from farmlands, unpaved roads and with only Labeo victorianus classified as critically footpaths. The site is also impacted by watering endangered. There was no clear dominance by a livestock, bathing and laundry by residents from the genus or species for the composition. Substrate was nearby Matongo town. Habitat conditions look good, variable and consisted of numerous substrate types, but there is potential for compromised water quality with gravel and cobbles being dominant. Habitat cover was generally dominated by aquatic vegetation. Habitat diversity was considered to be high, with three dominant velocity-depth class, namely FS, SD and SS. 4.4.2.6 Macroinvertebrates
124 because of mining. Table 4 38 and Table 4 39 below indicate the field processed and laboratory processed The increase in salinity during this period of slight score results for the site. rainfall increase could be due to salts which are 4.4.2.7 Water Quality washed into the river channel from the nearby High nitrate levels during the 1st survey possibly from agricultural areas. Runoff from these agricultural livestock manure when they water directly from the areas could also be the reason for the increased pH river. This level is just at the threshold recommended level. Slightly elevated levels of total nitrogen during for safe drinking water. Increased conductivity from the 2nd survey when there was slightly increased the initial to the 2nd survey contrary to expectations. rainfall compared to the initial survey. Table 4 40
shows the field measurements and laboratory results for water quality samples taken at the site.
125 126 4.3.3 Site Metrics
127 4.3.4 Indicators and Management Objectives
128 4.3.5 Required Conditions 4.3.6 Confidence Ratings local residents. The Tigithe River is dominated by a riffle-pool system, with fine gravels, silty clays, and 4.4 Tigithe large cobbles present. There are high terraces on The EFA biophysical site at Tigithe is located about 5 both banks and slight evidence of incision. There is km upstream from where it flows into the upper Mara commercial and artisanal mining in this catchment Wetland. It flows parallel to the Mara River and flows and there is concern over pollution in the river, directly into the Mara Wetland, near to where the although this is not related to alterations in flow. The Mara River enters the wetland. The area has higher fish here seem highly connected with the wetland rainfall than other areas of the Mara River Basin, species, indicating migration between the two although it is still considered a seasonal system by
FIGURE 4 31: SITE PHOTOS OF TIGITHE
129 areas. Flows during drought conditions are needed of water goes up to one meter. This has formed a to maintain good water quality in pools for aquatic wetland which its extension has been increasing over species. the years. 4.4.1 Social Survey 4.4.1.1 Lower Tigithe The Tigithe sub-basin consists of two RUs: the Lower Lower Tigithe has about 10 species of fish that were Tigithe RU and the Upper Tigithe RU. Tigithe River identified by the community, 17 species of natural is seasonal and dries up three times yearly from July vegetables, 17 species of fruit trees, 50 tree species to August. However, it is fed by many tributaries for building poles, five species of weaving reeds and which are either permanent or seasonal. Flooding seven species of thatching grass. Fourteen ecosystem in the Lower Tigithe RU mostly occurs in the lower services derived from wetland and rivers were and middle reaches of the rivers due to confluence identified to be useful for livelihood of communities of various tributaries. The inundation lasts for a day and extends up to three kilometers while the depth
in Lower Tigithe RU. Table 4 31 shows the wetland Matongo village (Table 4 32). Agricultural activities resources/ecosystem services obtained in this RU involve cultivation of crops such as maize, millet, and their relative importance to the two villages. bullrush millet, cassava, sweet potatoes, vegetables, Four main economic activities are practiced in this paddy/rice and bananas. The trend of resource RU i.e. Agriculture, livestock keeping (cows, sheep, utilization and condition of the wetland from the past chicken and donkey), petty trade, and mining in to the future is shown in Table 4 33 below.
130 131 in Lower Tigithe. This include two fish species (Mumi and Furu), 12 species of natural vegetables, 12 species 4.4.1.2 Upper Tigithe of fruit trees, 10 species of timber and building poles, The number of resources reported to be available in one weaving material and four types of thatching Upper Tigithe is much less compared to the number grass. The number of wetland resources and their
132 importance to the community is outlined in Table 4 34 below.
In terms of economic activities, only three were and condition of the wetland resources over the identified in the village: agriculture, livestock keeping years. A hypothetical value of 100 was set as a and petty trade (Table 4 35). current benchmark and perceived values of these Table 4 36 below shows trend of resources utilization
133 resources in the past and future with reference to the 4.4.2.1 Hydrology set benchmark indicated. The possible reasons for The Tigithe EFA site has a contributing upstream change have also highlighted. catchment area of 183 km2 with a dominated land 4.4.2 Biophysical – Tigithe River at Matongo use of agriculture, shrub land, and savanna (Figure 4
FIGURE 4 32: LAND USE WITHIN THE TIGITHE SUB-CATCHMENT (LEFT) AND CATCHMENT AREA (RIGHT)
134 32). Precipitation in the catchment sums up to 1,100 mm/yr, resulting in an evaporation value of 1,018 mm/yr, and a runoff value of 82 mm/yr. Average monthly, minimum and maximum discharge
FIGURE 4 33: AVERAGE MONTHLY, MINIMUM AND MAXIMUM DISCHARGE VALUES FOR THE TIGITHE EFA SITE values for Tigithe EFA site are presented in Figure 4 33. Typically, April is the wettest month and August along riffle sections. The hydraulic transect is shown is the driest month, going almost to zero m3/s. in Figure 4 34 with the observed and modelled flow 4.4.2.2 Hydraulics levels indicated. There was a weak relationship The Tigithe River follows a pool riffle sequence with between depth and velocity (R2 < 0.36), showing slow-flowing pools and steeper faster turbulent flow high variability in the velocity–depth association
135 FIGURE 4 35: FREQUENCY DISTRIBUTION FOR VELOCITY-DEPTH CLASSES FOR THE WETTED CHANNEL OF THE TIGITHE RIVER across the channel. The frequency distribution of the consists of silt and the floodplain/bench of layers of flow velocity-depth classes for the wetted channel are small gravel and sand. Recent flood deposited sand presented in Figure 4 35. is present on the floodplain/bench. The banks are 4.4.2.3 Geomorphology near vertical with active erosion along short sections Tigithe channel is a straight to sinuous channel with of bank. A narrow inset bench lines the right bank a pool-rapid sequence (Figure 4 36). The channel composed of medium sand. The riffles consist of slope is 0.007808 and can be classified as a channel armored cobble with voids filled with coarse sand of the upper foothill zone (Rowntree and Wadeson, and fine gravel. A silt drape is present on bed features 1999). Rowntree and Wadeson define the reference where the flow velocities are lower. The bed of the condition as “moderately steep, cobble bed or mixed pools consist of fine gravel and silt. bedrock-cobble bed channel with plain bed, pool riffle, or pool rapid reach types. The length of pools The Tigithe River plays an important role in and riffles /rapids are similar. Narrow flood plain of maintaining a large back swamp area to the North sand, gravel or cobble often present”. The site fits this of the Mara channel (Figure 4 38). The back swamp description well. forms on the Mara floodplain as a result of the alluvial ridge that forms due to sediment deposition along The channel is bedrock controlled and incised into the the Mara River. The Mara spills into this back swamp landscape as is evident by the narrow floodplain and terrace along the right bank (Figure 4 37). The terrace
136 Figure 4 36: A satellite image showing the transect position (white line) and the pool-riffle sequence along the Tigithe River (Google Earth Image date 17 Jan 2010).
FIGURE 4 37: CHANNEL CROSS SECTION FOR THE TIGITHE RIVER INDICATING THE GEOMORPHIC FEATURES (BLACK TEXT), SEDIMENT TYPE (BROWN TEXT) AND VEGETATION TYPE (GREEN TEXT).
137 FIGURE 4 38: A GOOGLE EARTH IMAGE SHOWING THE BACK SWAMP (ENCIRCLED BY THE WHITE LINE) FORMED BY THE ALLUVIAL RIDGE/LEVEE (BROWN LINE) ALONG MARA RIVER. THE TIGITHE RIVER DRAINS INTO THE BACK SWAMP PERMANENTLY AND THE MARA RIVER ONLY DURING FLOOD FLOWS. FLOW DIRECTION IS FROM EAST TO WEST.
during flood conditions, but the Tigithe permanently contributes to its water balance, making it a crucial brizantha. Grazing and browsing by livestock appear source of surface water during low flow conditions. to be high resulting in to lawns on the floodplain 4.4.2.4 Riparian Vegetation and pruned shrubs on terrestrial part of the system. The bed of Tigithe River is made up of cobbles stones Lantana camara was present in macro-channel bank. and sand and both left and right banks appear to be Overall, the vegetation cover was about 70 percent. stable. At its natural state the site would have more Figure 4 39 shows cross section and plant species vegetation cover in riparian area and at both banks. distribution at the site. Herbaceous vegetation would have been tall on the 4.4.2.5 Fish floodplain which is on left side of the river. There Tigithe River meanders with the flow in a westerly would also be more riparian trees on both river banks. and northerly direction for the reach surveyed. Two Absence of terrestrial tree species like Leucacena primary GHUs were identified and delineated for leucocephala and Lantana camara on the macro- the site, namely a pool and rapids (Figure 4 40). A channel bank would be expected. A strip of around total of four shallow pools and one deep pool were 100 m was selected to study riparian vegetation in sampled with an electro-fisher and fyke net for the Tigithe. study, with all pools characterized by slow velocity. Four rapid areas were sampled, with all these areas Members of Poacea and Cyperaceae families characterized by a FS velocity-depth class. Substrate dominated the marginal sub-zone including Cyperus in the pools is dominated by sand and cobbles, with cyperoides, Coix lacryma and Eriochloa macclounii. the substrate associated with the rapids characterized Lower sub-zone was occupied by Commelina by rocks, cobbles and bedrock. carsonii, C. benghalensis, Cyperus cyperoides and Cynodon dactylon. Among the studied sites this Turbidity was considered to be high. No aquatic was the site with more species richness in the Mara vegetation was present at the site, with marginal and basin. Floodplain area was present on the left side overhanging vegetation present for almost the entire and was dominated by grasses including Cynodon extent of the reach. Root wads and undercut banks nlemfuensis (around 60 percent), Setaria sphacelata, were also recorded for the majority of the pool units. Echinochloa sp., Hyparrhenia rufa and Brachiaria Table 4 37 and Figure 4 41 show an overview of the
138 139 140 HCR ratings. 4.4.2.6 Macroinvertebrates Some of sensitive macroinvertebrate taxa were collected at this site (i.e. > 2 species of Baetidae, A total of 15 fish species were sampled from the site, Tricorythidae and Hydropsychidae), and some that are with a total of 124 individuals recorded for the site. very sensitive to poor water quality (Heptageniidae, Two indicator species were recorded for the site, Leptophlebiidae). The site is affected by mining and namely Labeo victorianus and Clarias gariepinus, discharge of wastewater from mining, erosion and with only Labeo victorianus classified as critically sedimentation from farmlands, unpaved roads and endangered. There was no clear dominance by a footpaths. The site is also impacted by watering genus or species for the composition. livestock, bathing and laundry by residents from the Substrate was variable and consisted of numerous nearby Matongo town. Habitat conditions look good, substrate types, with gravel and cobbles being but there is potential for compromised water quality dominant. Habitat cover was generally dominated by aquatic vegetation. Habitat diversity was considered to be high, with three dominant velocity-depth class, namely FS, SD and SS.
141 because of mining. Table 4 38 and Table 4 39 below the initial to the 2nd survey contrary to expectations. indicate the field processed and laboratory processed The increase in salinity during this period of slight score results for the site. rainfall increase could be due to salts which are 4.4.2.7 Water Quality washed into the river channel from the nearby High nitrate levels during the 1st survey possibly from agricultural areas. Runoff from these agricultural livestock manure when they water directly from the areas could also be the reason for the increased pH river. This level is just at the threshold recommended level. Slightly elevated levels of total nitrogen during for safe drinking water. Increased conductivity from the 2nd survey when there was slightly increased
rainfall compared to the initial survey. Table 4 40
142 shows the field measurements and laboratory results
143 144 for water quality samples taken at the site 4.4.3 Site Metrics
145 4.4.4 Indicators and Management Objectives the Mara River flows into the upper Mara Wetland. 4.4.6 Confidence Ratings It is also located about 2.5 km downstream from the LVBWB river gauging station at Mara Mines. The 4.5 Mara Mines river is very wide at this location, with predominantly The Mara Mines EFA biophysical site is located sandy substrate (which becomes braided at low about 8.5 km downstream of the confluence with flows) and there is evidence of incision. The banks the Somoche River and 30 km upstream from where are very steep with a variety of terraces, making it
FIGURE 4.4.2: SITE PHOTOS OF MARA MINES
difficult for the river to overbank at this location. The surrounding area is dominated by rainfed agriculture flood is about 50 metres for the tributaries and up to and livestock grazing. two km for the Mara River with a flooding duration of 1 to 14 days and two months, respectively. The 4.5.1 Social Survey – Borenga, Gantamome depth of the water is about one half of a meeting in Two rivers are found in this RU, the Mara River and the inundated areas. A very diverse species of riparian Somoche River, which are fed by seasonal tributaries. vegetation and fishes (10 species) were identified in The area has large wetlands at the confluence of the the site. These provide the local communities with two rivers. These wetlands are important for locals’ natural vegetables (23 species), weaving materials, livelihood as most of the area is dry and mountainous hence not suitable for crop farming. Flooding occurs twice a year in April and December. The extent of
146 poles and thatching materials. The wetlands are important areas for crop farming and livestock keeping. Table 4 45 below shows the resources available and their relative importance in Mara Mines RU.
147 The two major economic activities performed in the villages are farming and livestock keeping, with
148 an addition of petty trading in Gantamome village (Table 4 46). Past and future trend of the wetland utilization and condition is indicated in Table 4 47 below.
4.5.2 Biophysical – Mara River at Mara Mines 4.5.2.1 Hydrology The EFA site Mara Mines has a contributing upstream catchment area of 11,283 km2 with a dominated
land use of grass- and shrub land (Figure 4 43). Average monthly, minimum and maximum discharge Precipitation in the catchment sums up to 954 mm/ values for Mara Mines EFA site are presented in yr, resulting in an evaporation value of 884 mm/yr, Figure 4 44. On average, the wettest month is May and a runoff value of 71 mm/yr.
149 and the driest month is August. The Mara River at pool sections do exist along the outer bends, creating Mara Mines typically does not stop flowing, but can deeper slower flow. In Figure 4 39 the channel cross reach very low flow levels depending on the rain section is presented with the observed and modelled patters in the MRB. flow levels indicated. 4.5.2.2 Hydraulics This section of the Mara River is characterized by a low During the low flow there was a weak relationship (R2 gradient sand bed river with low bed roughness and < 0.24) between depth and velocity, possibly due to a relatively square channel shape, creating relatively multiple channels of which some had very little flow. little hydraulic diversity. The hydraulic habitat is The modelled rating curve matches the observed low dominated by shallow slow flow during low flows and discharge and depth data closely, showing that the fast deep flows during high flows. During low flows hydraulic model performed well for the lower end of
FIGURE 4 45: CROSS SECTION FOR THE MARA RIVER AT MARA MINES FOR OBSERVED (DOTTED LINES) AND MODELLED (SOLID LINES) FLOW LEVELS
FIGURE 4 46: FREQUENCY DISTRIBUTION OF DEPTH-VELOCITY CLASSES FOR THE WETTED CHANNEL OF THE MARA RIVER FLOODPLAIN SITE NEAR MARA MINES 150 the rating curve (see Annex D for the rating curve). The river appears to be incised and is lined by levees Modelled velocity depth frequency distribution (Figure 4 48). The levees prevent overbank flooding classes are presented in Figure 4 46. at the site, limiting spills onto the floodplain to low points along the banks and flood channels. This is 4.5.2.3 Geomorphology evident by the narrow flood benches along both banks The Mara River at Mara Mines enters the floodplain with fig trees and shrubs lining it. Mature terrestrial and follows a meandering style with outer cut banks species are found on the flood plain, indicating and inner scroll bars (Figure 4 47). The river slope infrequent flooding. A cut-off meander depression is 0.0018 and is classified as a lower foothill river can be seen on the floodplain (top right of Figure 4 (Rowntree and Wadeson, 1999). This classification 47; ~1.5 m lower than floodplain surface), indicating is described as a ‘lower gradient mixed bed alluvial that the floodplain processes are still active. Outer channel with sand and gravel dominating the bed, bank erosion and scroll bar formation supports this locally may be bedrock controlled. Reach types finding. typically include pool –rapid/riffle with sand bars common in pools. Pools significantly longer than The banks are steep and composed of silt and fine rapid/riffle habitats. Floodplain often present’. The sand. The channel bed is dominated by coarse sand site does not fit the description well due to the lack with small superficial gravel patches. The bed material of rapid and riffle habitats, and long pools along the is loose with very little fine sediment trapped in the reach. interstitial spaces. The inset benches are very narrow along both banks and composed of fine sand and
FIGURE 4 47: AERIAL VIEW OF THE MARA RIVER AT MARA MINES SHOWING THE LOCATION OF THE CROSS SECTION, THE MEANDERING PLANFORM AND SANDY BED (GOOGLE EARTH IMAGE DATE 14 APRIL 2017).
151 silt. The inset benches along the right banks shows signs of erosion near cattle watering site. Local elders several sand bars and the river tended to meander remember a narrower tree lined channel from their through the sand bars. There are also flooding youth. benches which are dominated by grazing tolerant 4.5.2.1 Riparian Vegetation grass Cynodon nlemfuensis by almost 90 percent. The The site is expected to be dominated by more grasses riparian zone was covered with woody trees, shrubs and sedges at marginal and lower sub-zones which are and tall grasses especially the left side. Remarkably typical for riparian areas at its reference condition. Ficus exasperate dominated the left bank at lower Alien plant species are not expected at the two zones sub-zone. at reference. It expected that the vegetation in the upper sub-zone would be taller with more shade There was a large flooding area on the right side of resulting in less non-woody vegetation cover. There the river which is used for crop farming and grazing. would be large fig trees with intermediate class sizes Bank incision and slumping was also observed during suppressing establishment of alien plant species such the survey. Crop fields were also present on both as Lantana camara and Mimosa pigra. sides of the river and grazing pressure was high on both sides of the river as well. A shrub like Lantana The macro-channel bank is expected to be densely camara was present and provided refugia to many wooded and the proportion of terrestrial species is expected to be high due to infrequent flooding at reference condition. It was observed that the site had
152 forb species. There were few species at marginal sub- zone i.e. Urochloa brachyuran (grass) and Commelina The first reach consists of two dominant GHUs, benghalensis (forb). Figure 4 49 shows distribution of namely a rapid unit which is associated with the plant species along the cross section of the river. primary channel, and three runs (Figure 4 50). The 4.5.2.5 Fish water flow for these units is in a northerly direction, The Mara River at Mara Mine consists of two survey with all units characterized by a FS velocity-depth reaches. The first reach is located at the gauging class. The substrate for this reach of the system station and the second reach is further downstream; presents the delineated GHUs for the two reaches. The upstream reach was sampled with an electrofisher, and the downstream reach was sampled with a combination of an electrofisher, cast net and seine net.
153 was characterized by cobbles, rocks, boulders and the channel was dominated by stones and gravel, bedrock. Aquatic and marginal vegetation, root wads with root wads and overhanging vegetation present or overhanging vegetation were present. HCR ratings for the upstream reach is presented in Table 4 48. The downstream reach is classified as a shallow run with relatively fast flow (Figure 4 51). The reach is located in a meandering portion of the system, with sandbars present in the channel. The substrate within
154 although limited. Turbidity was considered high for Habitat cover also consisted of a variety of types, both the upstream and the downstream reaches. An with vegetation being dominant. The substrate for overview of the HCR ratings is presented in Table 4 the downstream reach was dominated by sand and 49 and Figure 4 52. gravel, with mud located on the channel edges. A total of 20 fish species were sampled from the site, The dominant cover was provided my marginal with a total of 99 individuals recorded for the site. Two indicator species were recorded for the site, namely Oreochromis niloticus and Clarias gariepinus. There was a dominance by Labeo sp for the composition. Substrate for the upstream reach was dominated by cobbles, rocks and bedrock.
155 FIGURE 4 52: A PHOTOGRAPH DEPICTING HABITAT CHARACTERISTICS FOR THE DOWNSTREAM REACH FOR THE MARA MINE SITE (FEBRUARY 2019) vegetation, with areas of root wads. Habitat diversity was considered to be high for the upstream reach and to highly sensitive to river flow and habitat low for the downstream reach, with two dominant availability (i.e., Naucoridae, Gomphidae, Lestidae, velocity-depth class, namely FS and FS. Baetidae, Simuliidae, Elmidae, Tricorythidae and 4.5.2.6 Macroinvertebrates Hydropsychidae), and some that are very sensitive Most of the macroinvertebrate taxa are moderately
156 to poor water quality (Heptageniidae, Oligoneuridae and Perlidae). Table 4 50 and Table 4 51 below indicate high pH could be an effect of geology much like the the field processed and laboratory processed score case in Tobora. The sandy channel could be playing results for the site. a big role in terms of cleaning/filtering the water 4.5.2.7 Water Quality hence the no change in turbidity observed during the The pH, conductivity and turbidity are all within the two surveys. Presence of nitrate was also measured same range during the high and low flows. The slightly
157 at this site possibly from the numerous number of
livestock that water directly from the river. Table 4 52
158 shows the field measurements and laboratory results
Annex J: Flow Setting Technical Meeting Report for water quality samples taken at the site.
161 4.5.3 Site Metrics 4.5.4 Indicators and Management Objectives mainstem Mara River. There is little habitat diversity 4.5.5 Required Conditions inside the channel since it is uniform in shape and 4.5.6 Confidence Ratings dominated by fine materials. There is a slight levee on 4.6 Bisarwi the banks, making them higher than the surrounding The Bisarwi EFA biophysical study site is located floodplain area. There is little vegetation on the banks in the upper Mara Wetland along one of the main and evidence of active erosion. The surrounding area flow channels on the northern side of the wetland. It is downstream of both the Tigithe River and the
FIGURE 4 53: SITE PHOTOS OF BISARWI
is predominantly flood-recession agriculture, with fields going right up to the banks. wetland vegetation i.e. papyrus. Since then, papyrus occupied the area causing the formation of Irirabo wetland. The wetlands have diverse species of fishes. 4.6.1 Social Survey – Kembwi, Marasibora The area is fairly in its natural state where fishing Initially, the area had no wetlands. These started to is done by approximately 60 percent of the locals, appear in 1997 during El Niño in which the water forming an important livelihood activity. Riparian overflowed to inundate areas where there were native vegetation are also in good condition as compared vegetation and trees. This suppressed the growth of native vegetation and trees causing the appearance of
162 to other sites. They provide locals with natural 57). vegetables, trees for fruits, poles and timber; weaving There are five main economic activities in the RU and thatching materials for subsistence use (Table 4 which include agriculture, livestock keeping, fishing,
163 164 small business and ulowaji (swimming) in Kembwi Village (Table 4 58). The trend of resource utilization and wetland condition is given in Table 4 59. 4.6.2 Biophysical – Mara Wetland at Bisarwi 4.6.2.1 Hydrology The EFA site Bisarwi has a contributing upstream catchment area of 11,903 km2 with a dominated
land use of grass- and shrub land (Figure 4 54). Precipitation in the catchment sums up to 960 mm/ yr, resulting in an evaporation value of 889 mm/yr, and a runoff value of 71 mm/yr.
165 Average monthly, minimum and maximum discharge values for Bisarwi EFA site are presented in Figure limited range of shallow flow habitats. This changes 4 55. On average, the wettest month is May and the dramatically once the river bursts its banks, flooding driest month is August. extensive floodplain areas that have shallow to deep 4.6.2.2 Hydraulics flow that could range from slow to fast. Figure 4 The Mara River at Bisarwi is typical of a low energy 56 shows the transect with observed and modelled system transporting fine sediment through slow flow levels indicated. Note that the channel fills first deep flows. The channel is narrow and deep with a before the water spills onto the floodplain. There
FIGURE 4 56: CROSS SECTION INDICATING OBSERVED (DOTTED LINES) AND MODELLED (SOLID LINES) FLOW LEVELS AT THE BISARWI EFA SITE
FIGURE 4 57: FREQUENCY DISTRIBUTION OF DEPTH-VELOCITY CLASSES FOR THE MARA RIVER AT BISARWI 166 was a moderate relationship between flow depth and velocity (R2 < 0.78). The velocity-depth frequency Local elders ascribe the avulsion to woody debris that distribution for the incremental increase in wetted caused a channel blockage during the 1989 floods. channel in Figure 4 57. A smaller channel has developed since, linking the 4.6.2.3 Geomorphology main channel just downstream of the site to the The Mara River at Bisarwi has a sinuous channel main channel that formed during the avulsion. It is pattern and is located one kilometer upstream of the anticipated that this new channel will increase in size main avulsion that took place in ~1989 (Figure 4 58). and become the new Mara River channel for the near The river channel slope is 0.000102 and is classified future. as a lowland river (Rowntree and Wadeson, 1999). They define the reference condition as a “low gradient The channel at the site is flanked by levees and is alluvial fine bed channel, typically regime reach type. deep and narrow (Figure 4 59). The upper banks May be confined, but fully developed meandering are steep and poorly vegetated, with active bank pattern within a distinct flood plain develops in erosion and bank slumping and no signs of sediment unconfined reaches where there is an increased deposition. Cattle accessing the river for drinking silt content of the bed or banks”. The site fits the adds to the trampling and degradation of the banks. description to a moderate level due to the cohesive The banks, bed and floodplain are composed of nature of the banks that resist lateral movement of silt. The floodplain slopes away from the levees and classic meandering channels. backwaters and flood channels are present leading floodwater away from the main channel (Figure 4
FIGURE 4 58: AERIAL VIEW OF THE MARA RIVER (BLUE ARROWS) AT BISARWI INDICATING THE LOCATION OF THE TRANSECT AND THE BIFURCATION OF THE MARA RIVER UPSTREAM OF THE AVULSION THAT TOOK PLACE IN 1989 (GOOGLE EARTH IMAGE 21 JULY 2017). THE BLACK ARROW INDICATES THE NEW CHANNEL AND THE WHITE ARROWS INDICATE FLOOD CHANNELS.
167 59). Higher ground on the floodplain has grass and shrubs growing, whereas the lower areas vegetated common species within floodplain were Sida acuta with grasses and sedges. Floating water plants and (invasive), Sida alba, Rorippa micrantha and Conyza rushes are present in the back waters. bonariensis (all forbs). Water hyacinth Eichhornia 4.6.2.4 Riparian Vegetation crassipes was also present but at low numbers. Hippo The site is expected to be dominated by more grasses grass, Vossia cuspidate, was present but only in few and sedges at marginal and lower sub-zones at places on the lower sub-zone. Figure 4 60 shows the its reference condition including a good cover of distribution of plant species along the cross section wetland plant species including hydrophilic grasses of the river. and sedges. It is also expected to have in-stream vegetation which includes macrophytes and fringing 4.6.2.5 Fish submerged aquatic plants that provide in-stream The Bisarwi reach of the Mara River has a sinuous cover and food for aquatic fauna. Only a few wetland/ channel pattern, flowing in a westerly to north- riparian woody plants are expected during high flows westerly direction. A total of three GHUs were due to inundation. identified and sampled for this reach (Figure 4 61). Sampling included a combination of electrofishing, During the survey it was observed that the site cast net and seine net efforts. These included a pool experience high human disturbance through cattle associated with the Mara River itself, and a run grazing and banks were unstable and slumping. Water flowing in a southerly direction from the Mara River. was very turbid indicating that clay soil erosion and The pool and the run were characterized with SD diverse human activities are taking place on the site and FS velocity-depth classes. A total of three pools, and up the river. There were papyrus stands about 50 all isolated from the Mara River were also sampled, m away from the river on right hand side. There were but yielded no fish species. These pools were shallow virtually no plants on the marginal sub-zone except with no flow (NF) presented. The substrate of the some dead grass species at the lower sub-zone which could be a result of prolonged inundation, denied moisture and/or effect of trampling. A large part of the marginal and lower sub-zones were bare. At the upper part there was 1 stand of Mimosa pigra on the right bank. The floodplain on the left side of the river was dominated by intensively grazed grass Cynodon nlemfuensis (60 percent). Other
168 169 FIGURE 4 61: AERIAL VIEW OF THE MARA RIVER SHOWING THE DELINEATED GHUS FOR THE SITE (GOOGLE EARTH)
Mara River and the tributary (run) was dominated by Two indicator species were recorded for the site, silt and clay. Aquatic and overhanging vegetation was namely Oreochromis niloticus and Clarias gariepinus. recorded for the Mara River. Table 4 60 and Figure 4 There was a dominance by Schilbe depressirostris for 62 show an overview of the HCR ratings. the composition. A total of 17 fish species were sampled from the site, Substrate for the reach was dominated silt, mud and with a total of 164 individuals recorded for the site.
FIGURE 4 62: A PHOTOGRAPH DEPICTING HABITAT CHARACTERISTICS FOR THE MARA RIVER REACH AT BISARWI (FEB 2019)
170 sand. Habitat cover also considered to be generally limited, with aquatic and marginal vegetation the reduced and most of the taxa are those that prefer dominant types. The reach of the Mara River was slow moving water/wetlands. Only Baetidae, among characterized by a SD velocity-depth class. the Ephemeroptera, Plecoptera, and Tricoptera taxa, 4.6.2.6 Macroinvertebrates were collected at the site, and none of the rheophilic This is a floodplain site where flooding occurs during taxa. The site is highly impacted by grazing and the rainy season, but during the dry season water in fishing. The floods have cleared all the vegetation confined to the main channel. Instream habitat is
171 along the river, a sign of high sediment loads coming from upstream. Table 4 61 and Table 4 62 below the drinking water standards in both periods of indicate the field processed and laboratory processed survey. This may be brought about by the presence of score results for the site. very fine sediments in the river channel which result 4.6.2.7 Water Quality in turbid waters. Increase in sediment loading in the The major difference in the parameter measured river channel and overland flow as a result of high during the two surveys was turbidity with high values rains was attributed to the high turbidity observed during the 2nd survey; but values having exceeded
then. Table 4 63 shows the field measurements and
172 laboratory results for water quality samples taken at
173 the site.
174 4.6.3 Site Metrics
175 4.6.4 Indicators and Management Objectives 4.6.5 Required Conditions The Mara Wetland EFA site is located in the most 4.6.6 Confidence Ratings downstream end of the wetland and is the most downstream EFA study site. It is about 15 km 4.7 Mara Wetland upstream from the outlet of the Mara River into
FIGURE 4 63: SITE PHOTOS OF MARA WETLAND
Lake Victoria. The inside of the wetland is dominated by floating papyrus, while the land on the edges of The wetlands are richest in terms of fish species (9 the wetland are predominantly raid-fed and flood species) when compared to other sites. Fishing is done recession agriculture. throughout a year and forms part of key livelihood 4.7.1 Social Survey – Ryamisanga, Wegero activities in the area. Other resources present include The Mara River, which flows throughout the area, natural vegetables (10 species), natural fruit trees (12 inundates the greater extent of the wetlands in South species), trees for building poles (15 species), weaving Mara RU. The wetlands experience the greatest materials and thatching grass. inundation twice a year in April and December. The Of all the resources available for livelihood, water maximum inundation reaches a distance of about was reported to be the most relative important and 18 kilometers in Wegero Village which is the largest accounts for an average of 80 percent. Livestock inundation coverage among all the studied RUs. pastures were the second important (7.5 percent)
176 followed by fish (6.9 percent) and wetland areas for farming (3 percent). Other resources accounted for less than one percent of relative importance (Table
177 4 68). The main economic activities include agriculture,
178 livestock keeping, fishing and small-scale businesses (Table 4 69). The trend of resource utilization and wetland conditions is given in Table 4 70 below. 4.7.2 Biophysical – Mara Wetland at Ketasakwa 4.7.2.1 Hydrology The EFA site Mara Wetland has a contributing upstream catchment area of 13,272 km2 with a dominated land use of grass- and shrub land (Figure 4 64). Precipitation in the catchment sums up to 985 mm/yr,
resulting in an evaporation value of 912 mm/yr, and wettest month and August being the driest month. a runoff value of 73 mm/yr. However, there are large influences from Lake Victoria and potentially inputs from groundwater, Average monthly, minimum and maximum discharge which ensures that there is water available year- values for Bisarwi EFA site are presented in Figure round. The influence from these other sources have 4 65. At this location in the wetland, the flow from rainfall runoff can vary widely, with May being the
FIGURE 4 65: AVERAGE MONTHLY, MINIMUM AND MAXIMUM DISCHARGE VALUES FOR THE MARA WETLAND EFA SITE
179 not been quantified so it is unsure how much of the Mara River influences the hydrological conditions at has shallow to deep stagnant water that could the site. become faster flowing during higher flows as there 4.7.2.2 Hydraulics is low roughness provided by the sparse vegetation. The Mara wetland is dominated by floating papyrus Observed and modelled flow levels are indicated in over very slow flowing to stagnant water with a depth Figure 4 66. Observed flow velocity and depth had a of 1 to 1.5 meters during the dry season. The open weak relationship in the wetland channel (R2 < 0.24). channel has deep slow flow. The edges of the wetland This is due to the relatively slow flow velocities across
FIGURE 4 66: CROSS SECTION FOR THE SOUTHERN EXTENT OF THE MARA WETLAND. DOTTED LINES INDICATE OBSERVED FLOW LEVELS AND SOLID LINES MODELLED FLOW LEVELS. THE GREEN RECTANGLES INDICATE THE EXTENT OF THE DENSE PAPYRUS PLANTS.
FIGURE 4 67: FREQUENCY DISTRIBUTION FOR VELOCITY-DEPTH CLASSES FOR THE MARA WETLAND 180 a range of depths. Figure 4 67 the modelled frequency distribution of the flow velocity-depth classes for the The hillslope forms a flood-prone area with grassy wetted channel at a range of inundation levels. vegetation that extends 2 m vertically above low flow 4.7.2.3 Geomorphology water levels (Figure 4 69). Large areas of papyrus is The Mara River forms a wide wetland (2 – 12 km burnt, but the burning seems to affect the above water wide) with an active channel along the southern stems only, leaving the rhizomes largely unaffected. margin. The channel is less pronounced in areas The flow velocity in the main channel is < 0.05m/s where papyrus forms a thick mat across the channel. and measuring flow under the floating vegetation The study site is located along a narrower part of the was not possible due to the thickness and density of wetland before entering Lake Victoria. The channel the vegetative matter. Small areas of coarse sand are at the site is ~ 60 m wide and the bed and flooded available where tributaries form alluvial fans along plain under the papyrus consists of silt (Figure 4 the margin of the wetland. 68 and Figure 4 69). The silt is compacted in the channel, compared to less compacted material along The water slope at the site is 0.0000255. If we extend the edges of the vegetation/papyrus. Shallow to this slope to the edge of the lake (roughly 16 km away) deep backwaters exist closer to the left bank, with we can estimate an elevation difference of 40 cm. If bedrock cropping out in-between the silty bottom. we extend this slope across the entire papyrus section
FIGURE 4 68: PLAN VIEW OF THE MARA WETLAND TRANSECT AND FLOW DIRECTION (GOOGLE EARTH IMAGE DATED 28 DECEMBER 2018).
181 FIGURE 4 69: CROSS SECTION OF THE LEFT BANK SIDE OF THE MARA WETLAND INDICATING THE MAIN GEOMORPHOLOGICAL FEATURES, SEDIMENT COMPOSITION AND VEGETATION TYPES. NOTE THE WETLAND EXTENDS FOR ANOTHER 2 KM TO THE RIGHT BANK. THE DASHED LINE INDICATE THE LIKELY BATHOMETRY OF THE CROSS SECTION THAT WAS NOT SURVEYED. of the wetland (35 km length), we can estimate an elevation difference of 86 cm. This would suggest that field survey it was observed that the site experience the Lake level has a significant influence at this site. minimal human activities. Water hyacinth was 4.7.2.4 Riparian Vegetation present in large numbers and the genus cyperus The site is expected to be free of alien and invasive (Cyperus glaucophyllus, Cyperus laxus and Cyperus plant species and be dominated with typical wetland papyrus) dominated the area by 95 percent. Of the plant species including hydrophilic grass, sedges three, Cyperus papyrus was the most dominant (70 and macrophytes at reference condition. Often times percent). Sesbania sesban a perennial legume tree plant distribution and composition follows mosaic was also present together with Nymphaea nouchali pattern but with majority (dominant) being wetland (fern) and Polygonum senegalense (herb). Floodplain plants particularly papyrus stands. formed part of the system on the south east side and it was dominated by intensively grazed short grass Woody plant species might be present at off-shore of Cynodon nlemfuensis which covered almost 60 the wetlands but not dominant. Annual herbs and percent of the floodplain. Beyond the floodplain there forbs are present. Many times species richness at is a clear line of trees which indicate that flooding is core wetland is not as high as in the floodplain and the composition between the two differs. During the
182 active and restrict encroachment of woody species particularly trees. Figure 4 70 shows the distribution dense stands of Papyrus (predominantly) could not of plant species along the cross section of the river. be access and sampled. One GHU was identified and 4.7.2.5 Fish delineated for the study, but with varying velocity- The Mara Wetlands are located in the lower portion depth classes (Figure 4 71). A SD pool was delineated of the Mara River Basin. The reach of the system for the main channel, with SS pools delineated on the considered for the study is associated with an active periphery of the wetland system. Sampling included channel along the southern margin of the system, a combination of electrofishing, cast net and seine flowing in a westerly direction. For the purposes of net efforts. The substrate for these units is dominated the study, GHUs have been delineated for areas of the by silt. Aquatic and overhanging vegetation was in system which could be accessed and sampled. As a result of this, areas of the system characterized by
183 abundance for the habitat units. An overview of the site, namely Protopterus aethiopicus and Clarias HCR ratings is presented in Table 4 71 and Figure 4 gariepinus. The fish diversity and abundances are 72 below. generally considered to be good. A total of 18 fish species were sampled from the site, with a total of 244 individuals recorded for the Substrate for the reach was dominated silt, mud and site. Two indicator species were recorded for the sand. Habitat cover also considered to be generally limited, with aquatic and marginal vegetation the
184 dominant types. The reach of the Mara River was characterized by a SD, and with the side water areas are inundated. The main channel has flowing water characterized by SS velocity-depth class. which provides riverine habitats for some sensitive 4.7.2.6 Macroinvertebrates taxa such as Baetidae, but no rheophilic taxa occur This is a wetland site that is flooded during the rainy at the site because of the slow water flow (low velocity season, with large areas of rooted macrophytes which
185 < 0.3m/s). Table 4 72 and Table 4 73 below indicate the field processed and laboratory processed score consumes most of the oxygen. Acceptable levels of results for the site. turbidity in the water during the 1st and 2nd survey. 4.7.2.7 Water Quality In as much as the turbidity was low, the water had Very low oxygen levels measured at the site during a light-brownish/tea color, possibly from high both sampling events. This could be as a result of organic load in the water resulting from the charred the slow flowing (almost still) water and presence of organic matter from decaying papyrus which
186 remains of burnt papyrus. Table 4 74 shows the
field measurements and laboratory results for water
187 quality samples taken at the site.
188 4.7.3 Site Metrics
189 4.7.4 Indicators and Management Objectives 4.7.5 Required Conditions should be placed on drivers of fisheries and 4.7.6 Confidence Ratings dependence and determination of conservation plan 4.8 Data Gaps for species. Some of the knowledge gaps identified by the experts - Historical diversity of the streams feeding into are outlined below: the lower Mara River need to be investigated. This should include taxonomic lists of major groups of - Where and at what discharge does flood water spill taxa, especially the most threatened (crabs) and flow onto the floodplain and how does this affect velocity- and habitat sensitive taxa (Oligoneuriidae, Odonata, depth habitat types across the surface? and other Ephemeroptera, Plecoptera, and Tricoptera taxa). - How does the floating papyrus influence the hydrodynamics of the wetland? - There is a need to investigate historical water levels (hydrology) and permanence of the streams feeding - What is the bed topography across the floodplain into the lower Mara River. All the three tributaries and permanent wetland? Are there preferential (Tobora, Somoche and Tigithe) are currently seasonal, pathways across the floodplain and wetland? How but it is possible that before the extensive land use dynamic are these preferential flow pathways and and land cover changes and intensification of human what is maintaining/threatening them? activities, including mining, in the catchments, these streams were permanent. This will have implications - How variable is the geomorphic template under on the recommendations being made for the natural and present day conditions? conservation of the streams. - To what extent is the river incision and avulsion - There is lack of historical data making it impossible linked to tectonic activity, catchment land use and to check the trend of water quality in comparison climate change? to the present state. The available information is quite limited and inconsistent in terms of dates of - To what extent do hippos maintain pool and riffle monitoring from one station to the other hence not habitats? possible to make a longitudinal comparison. - How does river avulsion influence the physical - How does flow levels/discharge vary with water habitat across the floodplain? quality? - How will the extent and the character of the - To what extent/rate does the riparian vegetation floodplain and permanent wetland change if the Lake cover change in different parts (or RUs) of the Mara Victoria water level drops by 2-5 meters? basin? - How will the extent and the character of the - How does denial/decrease/changing of flooding floodplain and permanent wetland change if a large influence the wetland/floodplain plant biodiversity? dam is constructed on the lower Mara River? - More information on the biology, ecology and fisheries of the study area is required. Emphasis
190 191 4. FINAL RESOURCE QUALITY OBJECTIVES AND RESERVE RECOMMENDATIONS
In other words flooding dynamics and plant these numbers, monitoring activities and an adaptive biodiversity relationships need further research. management cycle are recommended for inclusion as part of implementing the reserve. Presented below are the results of the Lower 5.1 Final Resource Quality Mara EFA, which include the final RQOs and the Objectives calculations for the reserve, which include basic 5.1.1 Kogatende human needs and environmental flows. The RQO The Kogatende EFA site lies within the Serengeti RU, statements are intended to be a guide as to how and the majority of this land is inside SENAPA. The resources should be managed now and into the largest stakeholder in this RU is SENAPA and local future, which are then accompanied by targets on tour operators. As such, the largest concerns at this how to meet those management conditions and site involve the wildlife and associated ecotourism in numerical indicators that should be met. The RQO the area. In general, the pressure on the ecosystem statements were reviewed by the EFA technical team from human activities were considered low by and their site metrics, indicator species (or indicator stakeholders, with the biggest concern coming from functions), and management objectives for each infrastructure construction or improvements inside site were incorporated into the process. The basic the park. There are impacts from the annual migration human needs and environmental flow values are but the impacts are considered natural. Maintaining calculated in m3/second and m3/day to align with a high level of resource quality is important as it both hydrological calculations and the on-going WAP supports various macrofauna (hippopotamuses, process. crocodiles, elephants, ungulates, etc.) which is the main driver of tourism and the local economy. The The environmental flow values are also accompanied current resource quality conditions were determined by descriptions of indicator functions for both low to be high for all categories (the exception being poor flows dry season, low flow wet season, and high flows, conditions for fish), indicating that the ecosystem freshets, and floods, as well as potential consequences was in an almost natural condition. Stakeholders and if those flows are not met. the EFA technical team indicated there was a slight degradation in conditions, particularly for low flows Results of this assessment apply to conditions in and water quality. They would like to see the area the basin today. As the basin continues to grow slightly improved to return to natural conditions. All in population, conditions on the ground change, of these considerations resulted in a management and more information becomes available, these class of A, or a near natural conditions where the values should be updated. Ideally, this should occur natural flow regime is to be maintained. This is every five to ten years to align with the schedule of reflected in the RQO statements for the Serengeti RU updating the WAP. To help inform the updating of (Table 5 1). The indicators and management objectives
192 193 from the EFA technical team (Table 4 6) were adapted to develop the RQO targets and numerical indicators most important since natural waterways are the (Table 5 2). main water sources for people and livestock. Overall, 5.1.2 Tobora the stakeholders and EFA technical team considered The Tobora EFA site lies at the outlet of the Tobora the RU to be in a moderate condition and thought RU, which is managed by the Tobora WUA. The that all resource quality elements were degrading primary land use in the sub-basin is rainfed and except for high flows. They would like to improve flood recession agriculture, livestock grazing, and a these conditions, but overall would like to balance small amount of fishing. The pressure on the system human activities with environmental protection in from human activities was considered moderate by a way which supports the sustainable use of natural stakeholders, with the activities with the greatest resources. These considerations were combined for a impact being livestock grazing, invasive species, management class of B, which is a somewhat altered sewage/solid waste, and the development around hydrological condition but with relatively small towns. These degrade the conditions of the resource impacts to the ecosystem. The RQO statements were quality elements through the destruction of the developed to reflect the desire for sustainable use, riparian zone, increased erosion, pollution, and allowing all resource quality element to be managed destruction of habitat. in a somewhat altered condition (Table 5 3). The The stakeholders considered the resource quality elements of low flows and water quality to be the
194 195 indicators and management objectives from the EFA purposes. Overall, the current conditions were technical team (Table 4 17) were adapted to develop considered to be moderate, but there was a range of the RQO targets and numerical indicators (Table 5 4). conditions when separated out by resource quality 5.1.3 Somoche element (low flows were considered to be in poor The Somoche EFA site lies at the outlet of the condition, but high flows, water quality, and riparian Somoche RU and is managed by the Somoche WUA. habitat were considered to be in good condition). The The primary land use is similar to the Tobora sub- stakeholders and EFA technical team considered basin, where rainfed and flood recession agriculture, most resource quality elements to be degrading livestock grazing, and aquaculture are the major (except for instream habitat). Overall, they wanted economic activities. According to stakeholders, to see the condition of the resource quality objective the pressure on the system was considered to be improve. These considerations were combined for a moderate, with the largest concerns coming from management class of B, which is a somewhat altered deforestation of the riparian zone, pollutants from hydrological condition but with relatively small small-scale gold mining activities, and degradation impacts to the ecosystem. The RQO statements were of the ecosystem from livestock. Low flows and water developed to reflect the desire for sustainable use, quality were ranked as the most important resource allowing all resource quality element to be managed quality elements due to the reliance on these water in a somewhat altered condition (Table 5 5). The sources for domestic, agricultural, and fishing
196 197 198 indicators and management objectives from the EFA technical team (Table 4 28) were adapted to develop about the impact this is having on water resources and the RQO targets and numerical indicators (Table 5 6). biodiversity. Again, low flows and water quality are 5.1.4 Tigithe considered to have the highest importance of all the The Tigithe sub-basin is divided into two RUs: Upper resource quality elements since it is the only source Tigithe RU and Lower Tigithe RU. This was done to of water for domestic and livestock use. Stakeholders align with the existing management structure in this are concerned about the impacts of mining activities sub-basin, namely the boundaries of the Upper Tigithe and toxic material in the river as well as pathogens WUA and the Lower Tigithe WUA. For the EFA, there which may affect human health. Stakeholders in the was only one site (Tigithe River at Matongo) which is Lower Tigithe RU considered the current conditions located within the Lower Tigithe RU. For the RQOs, of the resource quality elements to be moderate, there will be two sets of RQO statements; however, while the Upper Tigithe stakeholders considered only one set of RQO targets and indicators as well as their conditions to be poor. These conditions also one set of environmental flow values will be presented match the findings of the EFA technical team. All since it is expected that the needs of both RUs will be stakeholders thought conditions were declining and met with these recommendations. wanted to see them improve so they could continue utilizing important ecosystem services in the future. The Tigithe sub-basin contains many disperse human activities, with the main economic drivers being These considerations were combined for a rainfed agriculture, livestock grazing, and small-scale management class of B in each of the RUs, which is gold mining. Overall, stakeholders considered the a somewhat altered hydrological condition but with impacts from human activities to be moderate with relatively small impacts to the ecosystem. The RQO the biggest pressures coming from mining, livestock statements were developed to reflect a desire for grazing, deforestation, and the impacts from villages. sustainable use, allowing all resource quality element Plantations of eucalyptus trees are also becoming to be managed in a somewhat altered condition (Table more common and some stakeholders are concerned 5 7). The indicators and management objectives from
199 200 201 the EFA technical team (Table 4 42) were adapted to develop the RQO targets and numerical indicators livestock grazing and watering. Consistent with the (Table 5 8). other RUs, low flows and water quality appear to the 5.1.5 Mara Mines be resource quality elements that are most important The Mara Mines EFA site is located within the Mara to the local communities as the Mara River is the Mines RU. Currently, there is no WUA in this area largest local water source for domestic and livestock and water is managed by the LVBWB directly and use and it supports local fish populations. Current utilized at the village or individual level. There were conditions appear to be moderate but showing no stakeholders present at the RQO workshop for this signs of degradation in almost every category, and RU. The RQOs presented have been developed by the it is expected that local communities would like to EFA technical team using the social and ecological see the resource quality elements maintained for information gathered at the site and have been future use. These considerations were combined reviewed by project partners. The primary land use for a management class of B, which is a somewhat is rainfed and flood recession agriculture, livestock altered hydrological condition but with moderate grazing, and some subsistence fishing. The impacts impacts to the ecosystem. The RQO statements were on the system from human activities appear to be developed to reflect the desire for sustainable use, moderate with the biggest pressures being the removal allowing all resource quality element to be managed of riparian vegetation (including riparian trees) in a somewhat altered condition (Table 5 9). The due to farming and impacts on the ecosystem from indicators and management objectives from the EFA
202 203 technical team (Table 4 54) were adapted to develop water resource for domestic use, livestock grazing, the RQO targets and numerical indicators (Table 5 and fishing, and it is also important to maintain local 10). wildlife (hippopotamus). Stakeholders and the EFA Bisarwi technical team considered the current conditions The Bisarwi EFA site is located within the North of the resource quality elements to be moderate but Mara RU and is managed by the North Mara WUA. with a declining trend in condition. Their desire is for The main activities in the area are rainfed and flood conditions to improve so the system can continue to recession agriculture and livestock grazing. According be utilized for future use. These considerations were to stakeholders, the impacts from human activities combined for a management class of B, which is a on the ecosystem are slightly above moderate somewhat altered hydrological condition but with with the greatest pressures being from invasive relatively small impacts to the ecosystem. The RQO species (including water hyacinth and eucalyptus), statements were developed to reflect the desire for deforestation of native trees, activities from villages, sustainable use, allowing all resource quality element and livestock grazing. Again, low flows and water to be managed in a somewhat altered condition (Table quality were ranked as the most important resource 5 11). The indicators and management objectives from quality elements since the wetland is the most used
204 205 the EFA technical team (Table 4 65) were adapted to develop the RQO targets and numerical indicators (for domestic use, livestock grazing, and fishing). (Table 5 12). Stakeholders and the EFA technical team considered 5.17 Mara Wetland the current conditions to be moderate and overall The Mara Wetland EFA site is located within the South stable, but with a slight degradation in habitat in Mara RU and is managed by the South Mara WUA. recent years. There is a desire to improve the resource The main activities are rainfed and flood recession quality elements in the area to allow for continued agriculture and livestock grazing, similar to the rest use of wetland resources. These considerations were of the Lower MRB. The stakeholders find that there combined for a management class of B, which is a is little pressure on the system from human activities somewhat altered hydrological condition but with with only moderate concerns coming from invasive relatively small impacts to the ecosystem. The RQO species, livestock grazing, and burning of wetland statements were developed to reflect the desire for vegetation. All resource quality elements were ranked sustainable use, allowing all resource quality element as having low impact from human activities, but to be managed in a somewhat altered condition (Table almost all were ranked as having high importance to 5 13). The indicators and management objectives from the communities due to a mix of ecosystem functions (such as breeding and refuge sites for fish and wildlife and nutrient regulation) and services to humans
206 207 the EFA technical team (Table 4 76) were adapted to develop the RQO targets and numerical indicators ranged from 0.006 m3/s in North Mara RU to 0.018 (Table 5 14). m3/s in Mara Mines RU (Table 5 15). The values for 5.2 Basic Human Needs BHNs are based on a daily requirement of 25 liters/ The final estimates for flow requirements for basic person/day and are expected to remain constant human needs were calculated in units of m3/day throughout the year. The BHN values are based on and also m3/s to align with the environmental flow the RUs to align with the planning units used in the values. The basic human needs values for 2018 water allocation planning effort. Estimates for BHN
208 209 requirements were also calculated for the 5, 10, and 20 years to align with the water allocation planning applying this relationship to the other months. In this process (Table 5 16). way, the environmental flows across the water year, 5.3 Environmental Flows mimic the natural shape of the hydrograph. After Environmental flows are the amount of water at the preliminary values were determined at the Flow different times of the year needed to “protect aquatic Setting Technical Meeting, they were reviewed to ecosystems and to secure ecologically sustainable ensure the values aligned when looking at the entire development” (per the definition of the reserve in river system. Tanzania). For the Lower MRB, these values were determined through the RQO process (Section Social requirements have been incorporated into 3.2) and resulted in a management class of A (the the ecological motivations for each site and were highest level of protection) in Serengeti RU to protect not separated individually. In general, the local wildlife resources and the related ecotourism, and a communities rely mostly on rainfed agriculture management class of B (a balance between resource and livestock keeping for their livelihoods, and use protection and use) in the rest of the RUs where local fish, native fruits and vegetables, trees, and grasses residents rely heavily on the river for everyday needs. from the river ecosystem to for their daily activities. Using these management classes as a guide, the Some also use the river water to meet domestic water technical team determined the environmental flow needs, although shallow groundwater wells are more requirements to meet these objectives. commonly used in the area. Environmental flows are expressed as monthly low During the low flow conditions throughout the flow values (in m3/s) across one water year (October year, local communities require enough water in to September) as well as freshets and floods with the channel to support their domestic and livestock specific durations and timing requirements. These needs, but they prefer that the floodplain remain building blocks of the environmental flow regime free of water as it is where they farm and graze were set for each EFA study site for both maintenance their livestock. They also require enough water to years and a drought years. For the monthly low flows, maintain specific species of riparian plants for food the percent of average flow was calculated to provide or building materials. During floods, it is important context. For the monthly low flows, freshets, and the floodplain is inundated to replenish soil moisture floods, the magnitude (in cubic meters per day, m3/ and fertility for crops, maintain meadows for grazing day, and million cubic meters, Mm3) for inclusion in livestock, and fill local water sources (including the parallel WAP effort in the Lower MRB. ponds, swamps, and seasonal tributaries). The values for the environmental flow were developed High flows are also when many fish are caught, which through group consensus with the technical team are an important source of food. If the flow regime and project partners during the Flow Setting does not support these activities, it could greatly affect Technical Meeting. Typically, needs of the most the ability of these communities to obtain their basic sensitive indicator was selected as the environmental needs of water, food, and income. In drought years, flow, ensuring that the other less sensitive ecological it is particularly important to meet the flow values and social indicators would also be met. The values since the available water resources are reduced, decided upon during this meeting were monthly low increasing the pressure on the system by humans. flows for the driest month, monthly low flows for the These uses can be applied to all EFA study sites in the wettest month, and high flow events (such as freshets Lower MRB except Kogatende, which does not have and floods) that are critical for ecological and/or any communities in the surrounding area. social functions. The final values from the technical meeting are summarized in Table 5 17 and Table 5 18. The final environmental flow values set for each site are presented in the following sections, along The two low flow monthly values were then used with brief descriptions of their motivations and the to develop the monthly low flow requirements of the other months. This was done by calculating the percent of the environmental flow compared to the average flow of the two known months, calculating the linear relationship between the two, and then
210 211 consequences if they are not met. Full motivations for each EFA component can be found in the starter hippopotamus dung) from riffles, move gravels documents in Annex B through Annex I. through the system, and inundate flood benches; and 5.3.1 Kogatende floods are important for scouring bars, flood benches, At Kogatende, located on the mainstem of the Mara and inset benches while moving the bulk of the River, the dry season low flow requirements during sediment through the system. Floods are important a maintenance year are to ensure that the marginal for vegetation because they prevent alien species zone vegetation has enough access to water during from establishing, allow native species to recruit in the dry months. The vegetation in this zone includes the marginal and lower zones after flooding events, grass, sedge, and forb species. Dry season low flows and stimulate growth and reproduction in established provide important recruitment habitat for newly plant communities. The flooding also replenishes hatched fish individuals and rheophilic species, soil moisture and nutrients in the banks and flood while also allowing movement between different benches. Freshets are particularly important for habitats in the river channel and flowing tributaries. flushing organic matter and fine sediment, enhancing For macroinvertebrates, they are important for habitat suitability for annual and perennial plants. promoting suitable substrate and periphyton growth Higher flows are important for fish species because for scraper species, providing flow within riffles to they allow for connection with tributaries and support rheophilic species, maintaining good water inundate gravel bars, increasing habitat diversity, quality in pools and backwater zones, as well as species diversity, and migration between different submerged marginal vegetation required for specific habitats. They are important for flushing fines from taxa. stable substrate and maintaining good water quality conditions for sensitive macroinvertebrate taxa. They If the flows are not met, this will likely decrease the also reduce predation of macroinvertebrates from abundance of the marginal zone plant species at insectivorous fish since there are additional habitats the site as well as recruitment of newly hatched fish for fish to access. and vegetation-dependent macroinvertebrate taxa, threatening the survival of these species. The lack A lack of annual freshets and floods over a number of proper in-channel habitats may cause a loss of of years prevents the maintenance of existing habitat specific fish and macroinvertebrate species, such as and development of new habitats by preventing flow sensitive species that rely on highly oxygenated geomorphological scouring and deposition, which water in riffles. limits the growth of annual plant species and threatens the survival of moderately flow sensitive The wet season low flows are important for plant species. It also prevents the connections with inundating the lower zone vegetation and supporting tributaries, decreasing fish habitat diversity which seed germination and dispersal. They also provide could lead to a loss of critical fish communities. conditions for rheophilic fish species to feed in Lack of high flows also prevents the flushing of riffle and rapid habitats, allowing them to recover important macroinvertebrate habitat, which could from the low flow periods. Wet season low flows are be detrimental to rheophilic taxa and cause less also needed to provide habitats for spawning and sensitive taxa to dominate the system. Without access recruitment of indicator fish species. The wet season to additional habitats, insectivorous fish also predate low flows provide depths and velocities that will help heavily on macroinvertebrates in one area, changing flow sensitive macroinvertebrate taxa survive, while the community structure in that habitat. flushing out organic matter (including hippopotamus dung) and fine sediments from in-channel In general, drought requirements are about the habitats, which will help sustain sensitive EPT taxa survival of individuals during extreme dry periods. (Ephemeroptera (mayflies), Plecoptera (stoneflies), For fish and macroinvertebrates, it is critical that and Tricoptera (caddisflies)). pools in the river channel are maintained and refreshed (either through constant low flow or If the wet season low flows are not met, then there through freshets) so that necessary water quality will be a reduction in the germination and dispersal conditions are maintained. If these flows are not of the lower zone vegetation species. The flows also met, the abundance of many flow-sensitive species provide critical habitats for various life functions of could decrease. The wet month low flows and freshet fish species and may cause fish populations to become functions are similar to the maintenance year on a unsustainable. The lack of wet season low flows could decreased magnitude, but are particularly important also be detrimental to the recruitment and larval for the rejuvenation of aquatic species since many development of flow-sensitive macroinvertebrate of individuals may be stressed from the dry month taxa. And if the organic matter and fines are not drought conditions. In addition, these higher flows removed, this could smother algae that is the main maintain important marginal vegetation which is also food source for scraper macroinvertebrates. In critical habitat for many fish and macroinvertebrate addition, the accumulation of organic matter in pools indicator species. Floods are not particularly common and backwaters (from the lack of flushing) could cause in drought years but can provide critical refreshment anoxic or hypoxic conditions for aquatic species. and flushing of the system. When flows become too Regular high flows are important from a low, there is the possibility of extreme conditions in geomorphological perspective because they maintain specific water quality parameters, including increased gravel bars (which are important for fish spawning temperatures and low levels of dissolved oxygen. habitat); freshets scour organic matter (including
212 One issue specific to Kogatende is the presence of large numbers of hippopotamuses in the river. They While the Serengeti macrofauna (including crocodiles produce a large amount of dung, which impacts in- and hippopotamuses) are not a specific indicator, it channel habitat by filling in riffles and covering is expected that the ecological conditions provided stable substrates in organic material. At extreme low by the environmental flows will provide adequate flows or in standing water (like disconnected pools conditions for both habitat and food sources for the and backwaters), the hippo dung can also cause resident populations. In particular, ensuring the anoxic conditions. This adds to the importance of presence of pools during dry months that are deep maintaining proper flushing of the system through enough to hold these animals have been considered higher flows, freshets, and floods to scour out the during this process. dung from habitats, and maintaining low flows so they can replenish pools in times of low flows or The full results of the environmental flow values drought conditions.
213 214 at Kogatende are presented in Table 5 19, details on freshets and flow values are in Table 5 20, and For geomorphology, freshets help to move sand out of environmental flow values are graphed in Figure 5 1. gravel and cobble beds and flush riffles, which creates 5.3.2 Tobora important habitat for fish and macroinvertebrates; For Tobora, dry season low flows are important to high flows inundate and transport sand into flood support the survival of marginal zone vegetation, benches, and floods move gravels and turn cobbles including grass, sedge, and forb species. They (improving habitat diversity while also scouring also refresh pools, which are important refuge to vegetation and inset benches to form new habitat migrating fish species, and provide habitat for flow- for recruitment and colonization). These higher flow sensitive macroinvertebrate species. If the dry month events also trigger physiological changes in most low flows are not met, there will likely be a decrease annual plant species in the upper and lower zones in the abundance of the marginal zone vegetation, and help disperse seeds on higher parts of the bank. also decreasing the available habitat for vegetation- dependent macroinvertebrates. If the refreshed pools The floods often clear debris and promote germination are not available as refuges for fish, it may decrease of riparian tree species. Floods also provide ecological the biodiversity in the area and reduce populations of cues for fish spawning and increase connectivity indicator fish species. The lack of proper habitat may between habitats (mainstem to tributary, tributary reduce the abundance of flow-sensitive and water- to floodplain, etc.). In addition, they are important quality sensitive macroinvertebrate species. for restructuring macroinvertebrate communities and reducing predation pressure from specific taxa Wet season low flows are important to inundate and insectivorous fish. If higher flow events are not lower zone vegetation and support seed germination met (including freshets and floods), habitat diversity and dispersal in the system. These inundated plants for fish and macroinvertebrates will decrease are important habitat for vegetation associated because sand will remain in gravel and cobble beds taxa, supporting larval stages of odonates and true (increasing embeddedness) and fill in riffles, the flood bugs, and providing attachment sites for other taxa. benches will not be replenished and/or shifted, and Maintaining these flows are important for migrating the vegetation may encroach the system due to a lack species during wet periods to open up additional of scouring. The lack of high flows will also prevent feeding habitats. If wet season low flows are not met, annual species from germinating and, over time, the seeds from the lower zone vegetation species the system could change from being dominated by may not disperse and germinate, reducing the riparian trees to being dominated by shrubs. The lack abundance of these species. The lack of inundated of floods could prevent fish from accessing additional plant habitat for macroinvertebrates may result in a habitats and reduce abundance, while the lack of lowering of biodiversity. For fish, the lack of habitat macroinvertebrate habitat will be a disadvantage to for conditioning and recruiting may reduce species sensitive taxa. abundance and potentially cause species losses. 215 For the local communities, the importance of maintaining proper fish populations (and their food It is important to note that subsurface flow and/ sources) is high since 45 percent of the residents or groundwater are likely feeding these systems engage in fishing activities. In addition, the annual during dry periods, and there is evidence from the flood helps to bring back important moisture and local communities that the river flows year round nutrients to the agricultural lands close to the river in a normal (or maintenance year), even during dry and replenishes seasonal water bodies that, along months. As such, groundwater management is very with the Tobora River, are important sources of important in this catchment. If too much groundwater domestic water. is abstracted that is hydrologically connected to the river, it could prevent low flows from occurring At this site, there are no environmental flow values during maintenance years. for drought conditions. It is thought that the river frequently goes dry in drought years and the aquatic In the environmental flow recommendations below, species present during those conditions often survive there are some months where the environmental pools fed by springs or groundwater. However, from flow exceeds the average flow for that month. There the ecological studies, it appears that most fish species are significant uncertainties associated with the do not use these tributaries in drought conditions average flow found using regionalization during the and the resident populations of macroinvertebrates dry months since regionalization is based on rainfall- are less sensitive taxa which are able to survive in runoff relationships and cannot account for water standing pools. that comes from groundwater or springs. It is likely that the Tobora River system is fed by groundwater This suggests these tributaries are not critical for (either shallow subsurface flow or deeper aquifers) but species survival and maintaining biodiversity in the there have been no studies completed on this topic. In Lower MRB. In addition, when pools stand for too addition, there is no river gauging station at this site. long or become too small due to evaporation and lack As such, there is no historical data record to compare of replenishment, there is the possibility of extreme the regionalization results to see how closely they conditions in specific water quality parameters, match actual conditions. This further stresses the including increased temperatures and low levels need to establish a river gauging station at this site to of dissolved oxygen. There is also the chance of begin collecting data on the hydrological conditions concentrating pollutants from humans (such as in the Tobora River. Once a substantial data record fertilizers, pesticides, and raw sewage) to unsafe levels has been built (at least a few years of daily water for aquatic life. These conditions could become fatal level and flow data), then these EFA results should for any aquatic organisms living in these pools. For be revisited and revised as needed. See Section 6.3.1 these reasons, the EFA technical team decided that for more details on the uncertainties related to basin it is acceptable for this system to go dry in drought hydrology. years and no environmental flow values have been set. The full results of the environmental flow values at
216 217 Tobora are presented in Table 5 21, details on freshets Somoche also is impacted by large uncertainties in and flow values are in Table 5 22 , and environmental the average flow values during dry months, causing flow values are graphed in Figure 5 2. the environmental flow values to be much larger than 5.3.3 Somoche the average flow during these times. It is likely that The motivations and consequences for environmental the Somoche River system is also fed by groundwater flows at Somoche are similar to those found at Tobora. (either shallow subsurface flow or deeper aquifers) but Something specific to Somoche is the presence of rare there have been no studies completed on this topic. In vulnerable fish species. Maintaining refreshed pools addition, there is no river gauging station at this site. during the dry season low flows, access to riffles and As such, there is no historical data record to compare other feeding areas during wet season low flows, the regionalization results to see how closely they and providing access to additional habitats during match actual conditions. This further stresses the freshets and floods are all critical for the survival of need to establish a river gauging station at this site to these species. Not providing these conditions could begin collecting data on the hydrological conditions result in a loss of abundance, or in cases of prolonged in the Somoche River. Once a substantial data record extreme drought, extinction of these species. Local has been built (at least a few years of daily water communities use the Somoche River as an important level and flow data), then these EFA results should source of domestic water and some flood recession be revisited and revised as needed. See Section 6.3.1 agriculture. for more details on the uncertainties related to basin hydrology. Similar to Tobora, no environmental flow values are set for drought conditions at this site. The The full results of the environmental flow values
218 219 at Somoche can be found in Table 5 23, details on freshets and flow values in Table 5 24, and graph of lands close to the river and replenishes seasonal the environmental flow values in Figure 5 3. water bodies that, along with the Tigithe River, are 5.3.4 Tigithe important sources of domestic water. There are many similarities between Tobora, Motivations and consequences are similar between Somoche, and Tigithe as they are tributaries with maintenance years and drought years. If drought similar topography and geomorphology. As such, the conditions last too long, there is a chance of extinction motivations and consequences for geomorphology, of local endemic fish species. Due to water pollution riparian vegetation, macroinvertebrates, and water concerns in the Tigithe system from nearby mining quality are the same as described for Tobora. However, activities, it is also important to have regular freshets since it is connected directly to the Mara Wetland to flush any pollutants from standing pools, which are and not the mainstem of the Mara River, there critical refuge during a drought. While the hydrology are important differences in the fish communities is uncertain in the system, it is hypothesized that low present at the site. During the dry season low flows, flows come from subsurface and/or groundwater. As it is important that pools are regularly refreshed, and such, it is important to properly manage groundwater shallow riffles are maintained for resident fish. If dry resources to ensure these flows continue to contribute season low flows are not met, the available habitat for to the river system. Similar to Tobora and Somoche, resident species can impact biodiversity in the fish there are some months where the environmental community. The wet season low flows are important flow exceeds (sometimes by a large amount) the for providing feeding and refuge habitats for resident average flow for that month. There are significant and sensitive species of fish, as well as deep pools uncertainties associated with the average flow and other habitats suitable for wetland species of found using regionalization during the dry months fish. If the wet season low flows are not met, this since regionalization is based on rainfall-runoff could impact species’ ability to condition and recruit, relationships and cannot account for water that potentially having a long-term impact on abundance comes from groundwater or springs. It is unknown and species viability. Freshets are required to allow how groundwater is contributing to the system as the movement of migratory fish between habitats, in there have been no studies conducted on this topic. particular between the wetland and river. Freshets There is also no historical flow record in the Tigithe, are important for cueing fish breeding, especially so the actual hydrological regime is unknown. This moving wetland species up into the river. Riffle further stresses the need to establish a river gauging and rapid habitats are also needed for rheophilic station at this site to begin collecting data on the fish species. If these conditions are not met, critical hydrological conditions. Once a substantial data habitats for fish species may not be available and record has been built (at least a few years of daily water abundance and biodiversity may be reduced. For the level and flow data), then these EFA results should local communities, it is important to maintain proper be revisited and revised as needed. See Section 6.3.1 fish populations (and their food sources) since 20 for more details on the uncertainties related to basin percent of the residents engage in fishing activities. hydrology. The full results of the environmental flow In addition, the annual flood helps to bring back important moisture and nutrients to the agricultural
220 221 values at Tigithe can be found in Table 5 23, details on freshets and flow values in Table 5 24, and graph and Kogatende is that many local communities of the environmental flow values in Figure 5 3. utilize the Mara River at Mara Mines (whereas there 5.3.5 Mara Mines are no local communities at Kogatende). For these The motivations and consequences for the Mara communities, the importance of maintaining proper Mines EFA study site are similar to those presented fish populations is high since 45 percent of the local for Kogatende because they are both located on the residents engage in fishing activities. In addition, the mainstem of the Mara River and the sites provide annual flood helps to bring back important moisture similar ecological functions. However, crocodiles and and nutrients to the agricultural lands close to the hippopotamuses are not commonly seen at this site, river and replenishes seasonal water bodies that are and as such, their habitat needs were not considered important sources of domestic water. Water quality for this site. One ecosystem function that is specific during the dry months is also important as there is to Mara Mines is the movement of sediments. Since a chance of concentrating pollutants from humans the substrate is predominantly sand, it is important (such as fertilizers, pesticides, and raw sewage) to that higher flows, freshets, and floods move sand unsafe levels, causing the water to be unusable for and gravels through the system and scour the sand, agriculture (i.e., high levels of salts) and/or unsafe ensuring gravel bars are maintained and exposed. for consumption, causing outbreaks in water borne Otherwise sand will fill important in-channel gravel diseases. habitats. The full results of the environmental flow values at Another prominent difference between Mara Mines
222 223 224 Mara Mines can be found in Table 5 27, details on freshets and flow values in Table 5 28, and graph of important for macroinvertebrate diversity, as a wide the environmental flow values in Figure 5 5. range of taxa require access to water during their 5.3.6 Bisarwi adult stages. Floods also allow floodplain preferring Dry season low flows are set to support the survival species of fish access to inundated floodplains and of marginal vegetation. This marginal vegetation backswamp areas. provides attachment sites for vegetation associated Because the river banks have formed levees, the taxa. It also provides refuge habitat where wetland main way of inundating the floodplain is through and migratory species of fish can feed and recruit. distributaries. If the flood flows are not met, these If these conditions are not met, there may be a distributaries will not be activated and maintained, reduction in abundance of marginal vegetation, which will reduce the connectivity between the river sensitive and moderately sensitive macroinvertebrate and floodplain habitats. If floods are not met, itis communities, and indicator fish in the study area. possible that wetland areas will dry up, allowing During the wet season, low flows promote germination terrestrial plant species to encroach on wetland and of lower zone plant species. Inundation of marginal riparian areas. If the wetlands dry up, many taxa vegetation also promotes larval development for a would be lost in the inundated areas, predation variety of taxa. It also enables resident fish species to would increase in the main channel, and there would breed and recruit, and migratory species to access to be a change in the macroinvertebrate community habitat and tributaries upstream. If these conditions structure. A lack of access to the floodplain would are not met, there may be poor recruitment of lower also prevent some fish species from reproducing, zone plant species as well as fish species, potentially reducing biodiversity in the area. decreasing their populations. Many of the vegetation- associated macroinvertebrate taxa may also be During drought periods, dry season low flows reduced, potentially decreasing biodiversity in the provide enough water to support the survival of area. marginal zone vegetation, while wet season low flows provide water to both the marginal and lower zone. Freshets are important at this site because they Small floods provide enough water to propagate inundate the higher banks and reach the marginal various annual and perennial species. Without these vegetation. If freshets do not occur, it reduces the flows, the abundance of flow-sensitive and riparian chances for reproduction and recruitment of plant species may be diminished, as might the habitats for species. When this system floods, distributaries are vegetation-associated macroinvertebrate taxa. Dry activated along the main channel and backswamp season low flows provide refreshed pools for fish areas are inundated. This flooding is important to and macroinvertebrate species while wet season low maintain connectivity between aquatic, riparian, flows and floods provide relief from extreme drought and floodplain plant communities and provides conditions. It also provides habitat for some species to nutrient inputs and increases moisture in floodplain breed and recruit and other species to move upstream soils. Access to the wetland and backswamp areas is to find additional habitat. When flows become too
225 low, there is the possibility of extreme conditions in food sources) is high since more than 60 percent of specific water quality parameters, including increased the residents engage in fishing activities. In addition, temperatures and low levels of dissolved oxygen. the annual flood helps to bring back important There is also the chance of concentrating pollutants moisture and nutrients to the agricultural lands, and from humans (such as fertilizers, pesticides, and replenishes seasonal water bodies that, along with raw sewage) to unsafe levels, causing the water to the Mara River, are important sources of domestic be unusable for agriculture (i.e., high levels of salts) water. and/or unsafe for consumption. The full results of the environmental flow values For the local communities, the importance of maintaining sufficient fish populations (and their
226 227 at Bisarwi can be found in Table 5 29, details on freshets and flow values in Table 5 30, and graph of the wetland. The flows provide enough habitat for the environmental flow values in Figure 5 6. reproduction and larval development of a variety of 5.3.7 Mara Wetland macroinvertebrate taxa. If these flows are not met, The Mara Wetland EFA study site has specific it could inhibit development of sedge and grass conditions that make it untenable to set species, decrease the abundance and diversity of environmental flow values. Firstly, the hydrology of macroinvertebrate taxa, and stress fish populations the site shows evidence of being influenced by the Freshets and floods: level of Lake Victoria, potentially significant inputs from groundwater, as well as upstream inputs from 1 freshet at 4 m for 14 days (annual), 2 freshets at 5 both the mainstem Mara River and the Tigithe River. m for 10 days each in May and Nov (annual), 1 flood However, the exact influence each of these vary event at 4.8 m for 14 days (1 in 2 years) depending on location in the Mara Wetland, the time of day due to tides in the lake, and the time of year The annual flood should reach the tree line at the due to changes in river flows. In addition, different site. This will reduce encroachment of terrestrial hydrological sources may support satellite lakes vegetation as well as deposit silt and clay to replenish and small local wetlands located on the edges of the moisture and nutrients in the soil. The inundation main wetland. This makes it very difficult to quantify also connects important wetland and floodplain how much flow would be required from upstream habitats, which promotes the movement of plant rivers to maintain ecological and social functions in propagules over a large area, encourages germination the system. Secondly, the cross-section of the main and growth of woody plant species in the floodplain, Mara Wetland at the EFA study site is covered by and provides floodplain-preferring fish access to floating papyrus vegetation which limits access. This new habitats. These inundated areas also provide made it impossible to measure left and right bank good habitat for taxa of macroinvertebrates that boundaries of the cross-section or to know how much need aquatic environments during their adult stages. water and associated water velocities were under If the floods are not achieved, this would cause a the mat of papyrus. While the technical team made disruption to sediment and nutrient distribution in measurements at this site, many assumptions were the outer edges of the wetland, potentially changing necessary when developing the cross-section and the plant communities from wetland to terrestrial, associated hydraulic model. This reduced confidence altering flow patterns into the wetland from surface in this model to the point that the technical team runoff, and degrading the quality of wetland habitat decided it should not be used. It was decided that the for plants, fish, and macroinvertebrates. This could technical team would provide only depth estimates for include negative consequences on community the different hydrological components based on their composition and biodiversity. Maintaining fish field assessments, and that these depth estimates populations are important since 15 to 40 percent of would serve as recommendations for this site. the local communities engage in fishing activities. When more data can be collected on the hydrology and hydraulics of the site, additional analyses can Drought Year be completed to determine environmental flow Dry season low flows: Aug, Depth: 2.4m values. These depth recommendations do meet The motivations and consequences are similar to the conditions of the RQOs, targets, and indicators those in a maintenance year. In a drought year, the outlined in Section 5.1.7 since the ecological and dry season low flows also ensure good water quality social functions at this site are more dependent on to promote better growth of macrophyte species, inundation depths and time rather than flows. which are also important habitat areas for vegetation associated macroinvertebrates. If these flows are Maintenance Year not met, it could lead to a decrease in abundance of Dry season low flows: August, Depth: 2.7m macrophytes and the species they support. Dry season low flows support the survival and recruitment of flow sensitive plant species, including Wet season low flows: May, Depth: 2.8m macrophytes that provide necessary attachment sites The motivations and consequences are similar to for certain taxa of macroinvertebrates. These flows those in a maintenance year. In a drought year, also provide refuge habitats for resident wetland the wet season low flows also provide moisture to species and some migratory species of fish. If these sedge, grass, and forb plant species at the edge of the flows are not met, it could decrease the abundance permanent wetland area, supporting their survival of macrophyte species, impacting macroinvertebrate during dry times. It also opens up habitat to fish and attachment and young fish recruitment. This could macroinvertebrate species. If these flows are not met, lead to a decrease in macroinvertebrate and fish the plant community at the edge of the wetland could abundance. shift away from wetland species, impacting habitat for fish and macroinvertebrates. Wet season low flows: May, Depth: 3.3m Wet season low flows inundate the lower parts of the Freshets and floods: 1 freshet of 3.2 m for 14 days wetland and support the growth of sedge and grass (annual), 1 flood event of 4.4 m for 14 days (1 in 2 species. These flows provide habitats for resident fish years) to breed and recruit by opening access to inundated sedges and allow migratory fish enough water to The floods during a drought year have the same move upstream to other habitats in the interior of motivations and consequences, but to a lesser
228 geographic extent as the inundated area will be to a central database. When flow values begin to smaller. It also helps to refresh the standing water approach the environmental flow value, management in the wetland and flush out organic matter on the actions should be taken to reduce abstractions. edges of the wetland, promoting the germination of Specific values and actions will be determined in the sedge, grass, and forb species. It also would reduce WAP for the Lower MRB, which is being developed the chance of developing extreme hypoxic conditions by the MoW and the LVBWB. For effectiveness in the wetland. A lack of these floods could lead to monitoring, information should be collected to allow poor germination and abundance of sedge, grass, and for the regular assessment of ecological condition. forb species on the edge of the wetland and reduce This information can then be linked back to the the abundance of fish and macroinvertebrates. EFA management objectives and the RQO targets and indicators to see if the objectives are being met. 5.4 Monitoring and Adaptive Management Since the environment flow values were determined Monitoring for the Reserve using a variety of social and ecological indicators, a The reserve is an important value for effective variety of monitoring activities should be used.The implementation of water resources management monitoring recommendations presented here should in the Lower MRB, but it is not a static value. As be reviewed regularly by the LVBWB and updated conditions in the basin change, the reserve values to reflect their capacity and management priorities. should be updated regularly for both basic human To align monitoring activities to different levels of needs and environmental flows. For basic human institutional capacity (including financial and staff needs, the values should be updated using the capacity), a three-level system is proposed (Figure latest census information, projected to the current 5 7). Each level requires different levels of resource planning year. The amount of water should also commitments, including time, financial cost, and be updated to reflect any changes in legislation or necessary expertise. international best practice. For environmental flows, there are two main monitoring objectives: to ensure Level 1 techniques are non-technical and broad-scale, the environmental flows are being maintained in the with data that can easily be collected by a member river (compliance monitoring) and to ensure that of the public. Level 2 techniques are easily reported, aquatic ecosystems and important ecosystem services based on simple instruments, and data can be are being protected by the current environmental collected by a management authority who has a basic flow values (effectiveness monitoring). knowledge of hydrological or ecological processes. Level 3 techniques collect high quality and detailed Once these data have been collected, it is important data and are intended to be completed by experts in that they are incorporated into management decision the field. While monitoring can be collected at any through clearly defined adaptive management cycles. individual level, the data collected at one level should For compliance monitoring, it is important that contribute to the knowledge needed at next level (e.g., regular hydrological data are collected at monitoring basic information collected at Level 1 should provide sites. This can be achieved through automatic a foundation of knowledge for information collected monitoring equipment or regular observations from at Level 2). It is also not required to have monitoring local gauge readers who are able to send information activities at all three levels. The final monitoring plan
229 should be based on staff and partner agency capacity, financial resources, and the availability of monitoring For both compliance and effectiveness monitoring, methods available to collect useful data at each level. it is critical that a functioning river monitoring Implementing monitoring in the three-level system network be established. At minimum, water levels encourages collaboration with local partners. need to be measured at each EFA sites at regular Potential partners for monitoring include: intervals, such as twice daily or hourly. These water levels can then be translated to flow using regularly Level 1: WUA members, selected households (like updated site-specific rating curves. This allows the those engaged in farming, timber products, non- LVBWB to know if the reserve flows are being met timber products), village environmental committee, and if management actions need to be taken, such village health centres. as restricting abstractions by permit holders (the specific flows at which certain permit holders will be Level 2: Lake Victoria Basin Water Board, Musoma impacted will be determined in the WAP process). District Fisheries Department, district water The water level can be measured using a variety of engineer’s office, other local government authorities, methods, including automatic pressure sensors that WWF, Nature Tanzania, and other NGOs. can measure water levels at specific intervals and send the data back to the LVBWB as well as community- Level 3: The Universities in Tanzania including; based gauge readers (such as WUA members) that Sokoine University of Agriculture, University of Dar record and transmit information to the LVBWB at es Salaam, Tanzanian Fisheries Research Institute, specific times of the day. The exact data collection Ministry of Health, development partners. method selected should be based on the capabilities of the LVBWB and its partners, but it should provide As part of the EFA, technical team members suggested accurate, reliable, and timely information to the specific monitoring activities, including sites, timing, LVBWB. and frequency for monitoring, which are closely linked to the RQO targets and indicators (Table 5 The water level and flow information is important 31). These activities were suggested to help provide because the LVBWB needs to make important water more insight into the data gaps identified during management decisions to ensure the reserve is in the EFA process as well as ensuring the suggested compliance. To accurately assess how effective the EFA management objectives are being met. These reserve values are, the flows are also required since will need to be reviewed with the LVBWB and local the condition of the ecosystem and communities will partners to finalize which activities would provide need to be assessed against the flow record. If the information critical for management decisions and reserve is consistently met, then LVBWB managers which organizations would be best suited to carry out and technical experts can assess if the current reserve the activities. values are effective in protecting aquatic ecosystems 230 and ecosystem services. It can help assess if other a quality review conducted on incoming data. Being non-flow related issues are having a negative impact able to easily read, visualize, and sort data is important on the system. If the reserve flows are not met, it for ensuring it will be used in decisions making, and becomes difficult to determine if any decline in having a review process in place will improve the condition is due to lack of flow or other issues. quality and confidence in the data collected. This Maintaining a consistent and effective monitoring can be done using a simple spreadsheet program on program can be a challenge, especially when there are a computer or through free, online databases, but limited funds available and large distances to cover, the monitoring program leader should ensure it is but there are steps that can be taken to ensure that accurate and up-to-date. It is also important that the critical information is available to decision makers EFA information be gathered in a central location and when needed: combined with monitoring data from other LVBWB actions. It should not be a standalone database but First, it is important to assign a monitoring program rather integrated with other LVBWB information so leader who will be responsible for maintaining they can be easily compared and used in a variety of the monitoring network, organizing the gathered decision making activities. data, and ensuring decision makers have access to the database. This person should also act as a Fourth, when deciding what monitoring “champion” for the monitoring program, who will methodologies should be used, there are trade-offs apply creative problem solving as issues arise and that can be made in terms of complexity, reliability, ensure monitoring is considered a priority for fellow and expense. A simple monitoring method that staff members, managers, and budget holders. provides fewer data points but is more reliable may be a better choice than a complex method that collects Second, it is important to prioritize the types of data more data points but is prone to breaking down and collected. Data collection activities that are deemed expensive to repair. In addition, utilizing the WUAs critical (such as water levels) should be established and community members to gather simple data can first and monitored at all times, regardless of the be an affordable way to collect data across the basin financial situation. The collection of other data may without needing to travel long distances. need to be implemented in a phased approach or at a less frequent interval. While it may be difficult to And finally, it is important the budget holders make decide which data are critical and which can wait, it is it a priority to provide an annual monitoring budget important to establish institutional priorities which for these activities. While there are ways to collect are agreed upon by the monitoring program leader monitoring data in an affordable manner, there are and water managers. regular expenses that the monitoring program needs to pay in order to function, such as money for fuel, Third, it is important that the data collected are well- repairs to the monitoring devices, and payments organized and maintained in a central database, with for local assistants. The monitoring program leader
231 232 should prepare an estimate for the financial needs to maintain the monitoring program for the upcoming year and review it with the budget holder. and developing a water allocation plan (being 5.4.3 Incorporating Monitoring Data into completed in parallel to this effort). The next phase Adaptive Management is the reserve implementation phase. The first and A critical part of effectively implementing the reserve most important step in the phase is to complete is to incorporate monitoring directly into adaptive monitoring activities for both flow conditions and management cycles. This is achieved through regular ecological and social conditions. Regular monitoring monitoring (including data collection activities and should occur for both topics, but likely at different regular reporting) and the use of trigger values. time intervals (e.g., flow monitoring should happen These trigger values are numerical values for each continuously while ecological and social monitoring monitoring activity that indicate there may be a may occur seasonally or annually). During these condition of concern. It does not indicate that there regular monitoring activities, there are also periodic is a real problem, but it does trigger a management evaluation periods to analyze if the reserve flows action to investigate the issue in further detail. In this are being met (compliance monitoring) and if the way, major problems may be avoided by investigating objectives of the reserve are being met (effectiveness and/or taking action to alleviate the issue when it is monitoring). small. For the Lower MRB, the trigger values have been determined in the RQO indicators for each If the reserve is not being met, then different RU. Trigger values can also be updated using the management actions can be implemented to ensure information collected from the monitoring program that the required amount of water remains in the river, and utilizing the opinion of LVBWB staff and subject such as restricting certain water permits or reducing experts. the number of permits approved in subsequent years . If reserve flows are being met but the ecological and While adaptive management cycles should be social objectives are not, then it is time to evaluate the customized to incorporate existing management river system to determine the cause. If it is flow related, structures within the LVBWB, the adaptive then the reserve values may need to be adjusted. If it management for environmental flows could be based is not flow-related (such as impacts from changes in off of cycles recommended for the Rufiji Basin Water land cover), then alternative actions should be taken Board in central Tanzania (CDM Smith, 2018). These in collaboration with local government agencies and cycles separate activities for reserve determination development partners to alleviate the issue. If trigger (including updates every 5 years) and reserve values are surpassed during routine monitoring, then implementation. Implementing the reserve means the analysis phase is activated early (and potentially incorporating compliance monitoring into short- the next level of monitoring activities) to determine if term adaptive management action to ensure flows are there is a problem. being met in the river, and incorporating effectiveness monitoring into long-term adaptive management In addition to the monitoring, there are regular action to help determine if the environmental flow reporting periods. Hydrological data should values are achieving their intended objectives. These be complied, analyzed, and reported in annual cycles can be found in Figure 5 8. hydrological report sent to the MoW. Ecological and social data may need a few years of data collection The adaptive management cycle can be broken down before trends can be seen. These components should into different phases: the reserve determination be reported every five to ten years, comparing them phase and the reserve implementation phase. against the hydrological record and compliance The reserve determination phase includes setting record for the reserve over that same time period. initial reserve values (completed during this effort) These monitoring, management, and reporting cycles
5There will be years where the reserve is not met due to natural rainfall conditions. In these situations, it is recommended to use the drought year environmental flow values. However, if the reserve is not being met in average years, it is likely an issue of over abstraction of water from the river. 233 FIGURE 5 8: EXAMPLE OF ADAPTIVE MANAGEMENT CYCLES FOR ENVIRONMENTAL FLOWS FROM THE RUFIJI RIVER BASIN, TANZANIA (CDM SMITH, 2018)
234 235 6. DISCUSSION AND UNCERTAINTIES are completed continuously to ensure the reserve is and regulations, built upon previous knowledge, and being properly implemented and the correct reserve could be integrated into ongoing water allocation values are being applied. planning efforts. Next, authorities and stakeholders This report describes the process and results of an were engaged in a participatory process to set RQOs assessment to set RQOs and reserve levels for the for different RUs (aligned to sub-basins, Sections mainstem Mara River of Tanzania, its principal 3.2). These objectives guided the technical team in tributaries, and the wetland at its mouth. RQOs are the selection of targets and individual indicators for management objectives intended to protect water field assessments. Desktop studies were then carried resources and related aquatic biological resources out to quantify hydrological conditions across the at levels needed to meet the needs of resource users basin, classify ecosystem types and evaluate the level and maintain ecosystems in a desired environmental of flow alteration already influencing the river system management class. The reserve is a quantity of (Sections 3.3, 3.4, and 3.5). Based on these steps, water intended to i) satisfy basic human needs by the technical team, including counterparts from the securing a basic water supply and ii) protect aquatic LVBWB, selected seven sites for detailed biophysical ecosystems in order to secure ecologically sustainable field studies and 14 villages for a socioeconomics development and use of the relevant water resources. study, which were carried out during campaigns in RQOs and the reserve are recognized measures February and May 2019 (Section 3.6). Results of under Tanzanian law to protect water resources field studies enabled the technical team to relate and aquatic ecosystems. They are to be specified flow levels with the ecological condition of the river for all water resources in the country by notice of and the ecosystem services it provides (Section 4). the MoW in the Gazette. They are also to appear The final step in the process was a workshop of the as elements of a basin Integrated Water Resources technical team and water authorities to set reserve Management and Development Plan. The RQOs levels that meet RQOs for each site and to develop and reserve levels determined in this project comply a monitoring plan to support adaptive management with all requirements and approved guidelines under of RQOs and the reserve into the future (Sections 3.7 Tanzanian law and thus qualify for notice in the and 5). Gazette and application in continued water resource planning. They are judged to be valid for a period of RQOs set during the project reflect the close and five years, which corresponds to the validity period of multifaceted interdependencies of people and water the basin Integrated Water Resources Management and aquatic ecological resources in the Lower MRB. and Development Plan. After this period, and in People depend on river flows to meet water needs the context of regular water resource planning and for domestic purposes, livestock, and agriculture management, they should be reviewed and revised if across the basin. Groundwater is also an important judged necessary. source for domestic water. Special emphasis was given to dry season flows, but the importance of wet RQOs have been set for eight RUs (Figure 3 3), which season flows was also highlighted for supporting correspond in area to the six WUAs in the basin, floodplain agriculture and replenishing surface SENAPA, and the area around North Mara Mine. and groundwater storage for use in subsequent dry Reserve levels have been set for seven sites (Figure seasons. The importance of ecosystem processes is 3 9) aligned with the resource units and with Upper recognized as maintaining an ambient level of water and Lower Tigithe combined into one RU. RQOs have quality needed for healthy fisheries and water for taken into consideration water quantity and water domestic uses, livestock, and agriculture. Instream quality, the character and condition of in-stream and and riparian habitats and related biota are valued riparian habitats, and the characteristics, condition for the direct resources they provide (fish, building and distribution of the aquatic biota. Reserve levels materials, etc.) as well as their role in supporting have been determined for each month of the year, biodiversity. Biodiversity protection is recognized as as well as for years of normal rainfall and years of the predominate use for water in SENAPA but was also drought (Section 5). The monthly interval of reserve noted as important to inhabitants in all parts of the levels allows for combining months at seasonal or basin. These dependencies and values are recorded annual levels in planning processes. Values may also in the RQOs set by stakeholders and reported in be extrapolated to other points in the river system to Section 5.1. In all RUs of the Lower MRB, objectives align with different sub-basins or sub-units of water were set to maintain ecosystems in no less than a resource planning. As a next step in the process, the somewhat altered condition, which corresponds to reserve levels are to be extrapolated to the outlets of a class of B in the draft River Classification System sub-basins designated for water allocation planning. for Tanzania. In this class, the “natural flow regime The assessment was conducted by a multidisciplinary is affected by water withdrawals, impoundments technical team working in close cooperation with and/or discharges, but the critical aspects of the flow water authorities and stakeholders and following regime are retained so that effects on the ecosystem steps of the Nile E-flows Framework developed by are small.” the NBI and adopted by the riparian countries of the Nile, including Tanzania. First, a basin scale Results of the reserve assessment address the two situation assessment and alignment process (Section components of the reserve: the quantity of water 3.1) was conducted to ensure the project involved the needed to both satisfy basic human needs and correct authorities and stakeholders from national to protect aquatic ecosystems. The basic human needs local scale, met the requirements of Tanzanian laws component of the reserve is equivalent to 25 liters/
236 person/day and is thus directly related to population. the units used in this assessment can be reconfigured Based on the estimated population for 2018, this to match the final WAP planning units. The water amounted to flow levels of 9 to 40 liters per second balance of each planning unit can be summarized as depending on the resource unit (Section 5.2). The follows: ecological component of the reserve was set to meet Water Balance = Available Water – (Reserve + the RQOs and the environmental management class Transfers + Summation of Water Allocations) of B as described above. Specialists on the technical team set measurable targets and indicators related to A positive water balance indicates that there is the narrative objectives of the stakeholders (Section sufficient available water to meet all water demands, 5.1). These targets and indicators guided the flow while a negative balance indicates a state of over setting process. During years of normal rainfall, allocation. The water balance can be calculated at results for the environmental flow of mainstem Mara monthly, seasonal, or annual time intervals. The River sites corresponded to 23 to 24 percent of the results of this assessment quantified both the basic value of the average flow of the wettest month and 15 human needs and ecological components of the to 17 percent of the average flow of the driest month. reserve at monthly intervals, which allows them to be Environmental flows determined for mainstem sites incorporated into the water balance at whatever time during drought years were roughly 33 percent lower interval is chosen. It will still be necessary, however, than those for normal years. Environmental flows for to extrapolate reserve values to the outlets of final tributaries and the wetland correspond to larger or planning units. This can be done by adjusting the smaller proportions of estimated average monthly values reported in this document in proportion to the flow and even exceed the monthly average during upstream contributing area of each planning unit. the driest month. The more extreme proportions at these sites are predominantly due to uncertainties The reserve values are also relevant to the in the estimation of hydrological regimes, which management stage of water allocation planning, were regionalized from the relationships between including aspects of compliance and enforcement. precipitation data and mainstem hydrological In the implementation of the WAP, river levels are to records. be monitored to determine whether water users may continue to withdraw water at the full limit of their Environmental flows for normal years are intended permit or whether restrictions should be imposed to to support the full range of ecological process needed protect reserve flows in the river. The Draft Tanzanian to maintain healthy plant and animal communities WAP Guidelines recognize three levels of flow that in the river system. Detailed ecological motivations are relevant for water resource management; flood for each determination are given in Section 5.3. flow, normal flow, and reserve flow. Flood flows are Protection of ecological processes in the river also flows above an established threshold (e.g., Q80) that ensured continued delivery of ecosystem services represent a condition of abundant flow. Under these beneficial to human communities. Environmental conditions all water permit holders are expected to flows for drought years are intended to sustaining life be able to withdrawal water up to the limit of their in the system until higher flow levels return. permits. Normal flows represent flow levels below the flood flow threshold but greater than the reserve 6.1 Implementation of RQOs and flow level. Under normal flow conditions withdrawals Reserve Flows in Water Allocation may be restricted for some permit holders, such as Planning large-scale irrigators. Other permit holders such as As indicated above, the RQOs and reserve levels domestic water providers are expected to be able to determined in this project comply with all withdrawal water up to the limits of their permits requirements and approved guidelines under during normal flow conditions. When flow levels Tanzanian law and thus can be applied in water in the river drop to reserve levels, all water permit resource planning. Both are relevant for the water holders must cease abstractions, except for domestic allocation plan currently under development by water providers. However, even domestic water the LVBWB and the MoW. According to the draft providers should restrict their withdrawals to the guidelines for water allocation planning developed basic-human-need level of 25 liters/person/day. by the MoW, water resources allocation is “a means by which regulation of water use is done through In the water allocation planning activities planned sharing water resources among competing users, for the concluding months of 2019, RQOs and reserve with due regard for the environment, the economy, flows should be considered and used as indicated and the social wellbeing of all Tanzanians” (URT, above. 2018a). Setting RQOs is a required step in this process and a 6.2 Harmonization of Reserve Flows mechanism to incorporate stakeholder interests and with those set for Kenya align them with the requirements of Tanzanian laws Current estimates are that 75 percent of the water and regulations. During the planning stage of water flowing in the Mara River in Tanzania comes from allocation planning, a water balance is to be quantified Kenya. Thus, close coordination is necessary between for individual water bodies of planning units. In this the countries in water allocation and management. assessment, eight resource units were delineated to This also applies to consideration of the reserve. serve as the basis for setting RQOs and potentially as Fortunately, Tanzanian and Kenyan water laws planning units for the WAP. If alternative planning are consistent in their definition of the reserve and units are delineated during the WAP development, assigning it highest priority in water allocation. Both
237 countries include basic human needs and ecosystem hydraulic model used to convert ecologically relevant protection as components of the reserve. Both hydraulic variables like water depth, velocity, and countries recognize the basic human need to be 25 wetted width into discharge values. These hydraulic liters/person/day, and both countries have adopted variables were then linked to requirements of aquatic the Nile E-Flows Framework for the determination and riparian plants, fish, macroinvertebrates, and of the ecological component. This consistency in social uses. Flow levels were low during both field laws, definitions, and approaches greatly enhances campaigns, which means that the hydraulic model the potential for harmonious management of water is better calibrated for low flow conditions. This is resources across the border. Care must also be important because aquatic ecosystems of the Mara taken that numerical values of reserve flows and are currently most vulnerable to flow alteration implementation measures are consistent in a manner during low flow conditions. There is considerably that ensures Kenyan reserve flows crossing the more uncertainty in the performance of the hydraulic border are sufficient to meet Tanzanian reserve flow model at high flow levels, but this is less of a concern levels. Kenya also recognizes three levels of flow in because high flow levels in the Mara are largely the allocation and management of water resources, unaltered and are expected to remain unaltered namely flood flow, normal flow, and reserve flow, during the period these reserve determinations and uses a similar process of restricting withdrawals remain valid. based on flow levels. The environmental management objectives of Tanzania and Kenya at the border are The lack of long-term hydrological data for the similar given the juxtaposition of Serengeti National Somoche, Tobora, and Tigithe tributaries is of Park and Maasai Mara National Reserve. This should concern for water allocation because of the high lead to similar determinations of the ecological uncertainties associated with the regionalized data component of the reserve. The reserve determined in from the water resource assessment. This is especially this assessment at Kogatende in Serengeti National apparent in systems where groundwater or springs Park is judged sufficient to meet downstream reserve may have a substantial contribution to river flows requirements in the five year time period these during dry months since regionalization is unable determinations will remain valid. to capture these sources. There is some anecdotal evidence of this limitation that was encountered 6.3 Knowledge Gaps to Address during the field campaigns, where local residents Uncertainty is inevitable in any scientific assessment said the Somoche, Tobora, and Tigithe Rivers rarely of reserve levels, especially in data scarce systems go dry, which contradicts the no-flow values found in like the Mara River Basin. This assessment has been the regionalization for these three sites. transparent in acknowledging uncertainties and taking steps to minimize risks associated with them. This leads to uncertainty in the total quantity of The assessment team stands behind the reserve water available during different months of the year flows reported here but also strongly recommends and between different years. So, while there is higher that actions be taken to improve knowledge and confidence in the reserve flows, the uncertainty in understanding of key components resource system. the total water available is transferred to the water balance and volume of water available for allocation 6.3.1 Basin Hydrology for uses like domestic, livestock, irrigation, and Urgent action is needed to restore the industry. If regionalized data overestimate the total hydrometeorological monitoring network of the water available, this could lead to over-allocation of Mara River Basin. There are currently no functional water in permits and an increased risk of not meeting river discharge or precipitation stations in the basin. the reserve. If regionalized data underestimate the In this assessment, suitable historical data were total water available, this could limit the ability available from only one river discharge station (Mara to approve permits and unnecessarily limit the Mines) and two precipitation stations (Nyabassi and utilization of water resources. It is important that the Mugumu) which are near but outside the basin. Mean managing water authority has the confidence in the monthly river flows for all assessment sites other than water balance and the amount of water available for Mara Mines had to be reconstructed as explained allocation when making such decisions. in Section 3.3.1. Similarly, high flow values had to be simulated as explained in Section 3.3.4. Almost 6.3.2 Wetland Hydrodynamics nothing is known about groundwater, which was The hydrology and hydraulics of the Mara Wetland not explicitly considered in the reserve assessment. also remain largely unknown. During the field Daily precipitation and flow data are necessary for assessments the team measured flows in the wetland implementation of the reserve, and long-term data that significantly exceeded flows into the wetland at sets are necessary for proper planning of water Mara Mines. This indicates that flows in the wetland resource use and allocation. included drainage of stored water as well as inflows from the Mara River. Water levels in the lower The lack of historical hydrological data had a portions of the wetland also appear to be influenced minimal impact on reserve flows determined in this by level of Lake Victoria, which diminishes the degree assessment because the modified building block to which these portions of the wetland are dependent method used is based primarily on data collected on Mara River flows. Improved knowledge of these during the assessment itself. Accurate river discharge hydraulic characteristics as well as bathymetric data measurements were made during the two field of the wetland will strengthen the connection between campaigns and these data were used to calibrate the the hydrology and the plant and animal communities,
238 which is required to set appropriate reserve levels in wet months and extended reduced flow during dry Mara River in order to maintain sustainable habitats months (Mango et al., 2011), but it is not expected in the Mara Wetland. to have a significant impact on average annual flows 6.3.3 Low-Flow Ecology in the MRB (Roy et al., 2018; WWF-Kenya, 2019). Because the Mara River is presently most vulnerable With the potential for an increase in temperature to flow alterations under low flow conditions, there and a decrease in flow during dry months, there is is urgency to improve knowledge of how aquatic the possibility for water quality issues to arise during ecosystems function during low flows, especially these periods. strategies employed by river and riparian organisms to cope. Low flows are a natural part of the river’s There are some potential impacts on the indicators hydrograph and riverine and wetland species are used in the EFA. An increase in temperature causes a adapted to cope with natural low flow conditions. But decrease in dissolved oxygen in surface water, which, the increasing water demands of basin inhabitants if extreme enough, could impact habitats for aquatic during dry periods are likely to reduce river flows to species (USAID, 2017). In addition, the low water unnatural levels and to extend the duration of low levels may concentrate nutrients in the system (both flows. This will increase stress on river organisms to natural and human-introduced) to an unsafe level. A levels beyond which those organisms are adapted to further decrease in flow during dry periods may also be cope. of concern, although the system regularly experiences drought conditions and many of the aquatic species During the field campaigns, the mainstem Mara have adapted to these conditions. It is important that River flows were less than one m3/s which is near the mainstem Mara River maintains refuge habitats the minimal flows recorded historically in the river. during the dry months in both maintenance and Given these low flows, the technical team was drought years. These refuges were available during impressed by the abundance of organisms found and our field campaigns, where we encountered extreme their apparent good condition. But many questions low flows. While individual drought years followed remain about what undetected impacts may have by maintenance years appear to refresh the system been present, how much stress the organisms were properly, multiple years of drought conditions could under, and what the consequences would have been have a severe negative impacts on the population for of extending this stress. The adaptations of human species that migrate for reproduction (like fish) and communities to more severe and extended low flows those that require inundated soils (like vegetation). is also an area in need of additional study. The Tobora and Somoche tributaries were found to be important areas to maintain biodiversity during 6.4 Special Considerations maintenance years, but were not critical refuge 6.4.1 Climate Change habitat during drought years. Contrary to these, the Several studies have been conducted on how climate Tigithe tributary was important refuge habitat during change may impact water resources within the Lower drought years for specific wetland species, showing MRB (Mango et al., 2011; Dessu and Melesse, 2012; its vulnerability to a further decrease in flow during URT, 2014; Roy et al., 2018; USAID, 2019; WWF- dry months. The tributaries are also a primary water Kenya, 2019). While climate projections vary based source for many people living in those sub-basins on the climate models and scenarios used, in general, and are critical for providing water for basic human the expected impact is a 1.0 to 2.0 degree Celsius needs in those areas. increase in average temperature by 2030, a 1.5 to 2.7 degree increase by 2050, and a 3.5 degree increase To increase resilience to the impacts of climate by 2100. Changes have already been seen in the change (or decrease the potential for negative MRB, with an increase of 0.9 degree Celsius in the impacts from the expected changes in climate), average maximum temperature and an increase of 1.1 developing and implementing environmental flows degree Celsius in the average minimum temperature is considered an important step in both the Kenyan between 1961 and 2014 (USAID, 2019). An increase and Tanzania sides of the MRB (USAID, 2019; WWF- in temperature is likely to cause additional or Kenya, 2019). By including environmental flows in lengthened periods of water stress during dry established water management practices, the water months and more frequent and intense drought required to maintain aquatic ecosystems and their events. Annual average precipitation is also expected ecosystem services will be considered first priority. to increase, with approximately a 15 percent increase The ecosystem services will help to regulate extreme during the wet periods and almost no change during water events through the presence of riparian the dry periods, likely increasing the number of vegetation (increasing soil infiltration and slowing extreme rainfall and flooding events. Much of this surface runoff), contribute to food security through additional rain will occur during already wet periods, the maintenance of fish populations and indigenous resulting in surface runoff which, when combined fruits and vegetables, and provide enough water for with changes in land use, is likely to move quickly to the basic human needs of dependent communities. rivers and other surface water sources rather than There some impacts already being noticed in the infiltrate as groundwater recharge. This is also likely Lower MRB, such as the drying up of small streams to increase erosion as well as turbidity in the water. and some wetlands over the past decades (USAID, In addition, the increase in rainfall combined with 2019, Annex B). However, it is unclear if this is due the increase in temperatures is expected to increase to climate change or the more immediate impacts evapotranspiration. These conditions create the from changes in land use types. Converting the native potential for larger but shorter hydrological peaks in forest cover in the Mau Forest in Kenya (which form
239 the headwaters of the Mara River) to agriculture reports of persistent skin irritation and other ailments or grasslands has been shown to cause an increase in the area, which are thought to come from pollution in peak flows and a decrease in low flows in the from mining activities. Mara River (Mango et al., 2011). In addition, some There have been a series of studies on heavy metal farmers in the tributaries of Lower MRB are growing and mercury contamination in the Lower MRB with plantations of eucalyptus trees species as a source of varying results. Some studies conducted found no building poles and fuel, which have the potential to or only traces of heavy metals in Tigithe River and consume a large amount of groundwater and impact Mara River water samples (GLOWS-FIU and WWF- local streamflow in arid and semi-arid lands (Scott ESARPO, 2007; Almås and Manoko, 2012; Mataba et and Lesch, 1997). al., 2016), while others found that levels of chromium, cadmium, nickel, iron, mercury, and lead were all The time scale of the EFA is intended to be five to ten above Tanzanian Bureau of Standards and World years and should be updated along with the WAP for Health Organization standards (Bitala, Kweyunga the Lower Mara. In 10 years from now, there are not and Manoko, 2009; Kihampa C and Wenaty A, 2013). expected to be significant changes in the hydrology of One study found that levels of heavy metal were 14 Lower MRB. Using the BBM, the environmental flows to 260 times higher than when they were measured recommended are not based on the basin hydrology in 2002, a time before mining was a major industry but rather on the ecological and social needs which in the area (Bitala, Kweyunga and Manoko, 2009). are linked to habitat conditions (depth, velocity, Arsenic was not found in any of the river samples and inundation period). Once these habitat values in the studies since it breaks down quickly in open have been established, they are converted to flow waters. using site-specific hydraulic models and compared to known hydrology (measured or modelled) in the It should be noted that surface water quality changes basin or sub-basin. (This does, however, stress the rapidly and any collected samples are representative importance of an accurate hydraulic model and of the water quality at that specific moment. If there good hydrological data since the final flow values are were consistent leakage into the surface water, there dependent on these data sources. There is also the many a chance of collecting it in a sample. Often, possibility of updating these values using improved however, spills are individual events and are unlikely datasets and models as they become available in the to be captured by periodic studies. future.) It is important to remember that the reserve is one component of integrated water resources Sediments can capture pollution from longer management, and any decisions related to how water time scales since heavy metals in water often is allocated, including in response to any expected bind to sediments when moving through soils in changes in water availability due to climate change, groundwater and when in contact with sediments in will be decided in the WAP. surface waters. Contaminated sediments can also be deposited on land from local flooding events. Several 6.4.2 Heavy Metal Pollution from Mining studies found elevated levels of heavy metals in soils Activities immediately surrounding large-scale and artisanal In the Lower MRB, there are two main gold mining mining activities, indicating that there is likely some activities, each of which has its own impacts on water movement of polluted groundwater from these quality: large-scale mining (North Mara Gold Mine), activities (Bitala, Kweyunga and Manoko, 2009; which uses cyanide for extracting gold from raw ore, Mganga, Manoko and Rulangaranga, 2011; Almås and artisanal mining, which often uses mercury for and Manoko, 2012; Kihampa C and Wenaty A, 2013; the same purpose. For large-scale mining, the use of Mataba et al., 2016). While the heavy metal levels in cyanide chemically unbinds heavy metals (including the sediments are higher in concentration than the arsenic) from rocks and soils and results in a highly levels found in the water, the heavy metals are often toxic liquid known as acid mine drainage. While bound to sediments and may be less bioavailable, efforts are typically made to prevent acid mine decreasing the risk to living organisms (Ikingura drainage from leaving the mining area, it is capable et al., 2006). Sediment cores in the upper Mara of contaminating groundwater through seepage in Wetland show an increase in mercury deposits in the soil and contaminating surface water through the 1960s that are about 2.5 times the background overland spills. There is concern in the Lower MRB concentrations (although still well below the limits that there is inadvertent pollution into the Tigithe set by the US National Oceanic and Atmospheric River and surrounding ecosystem from the mining Administration’s effects range low concentration), activities at the North Mara Gold Mine. but have decreased to almost background levels in the past 20 years. The middle wetland is seeing an The main impacts from artisanal mining is the release increase in mercury deposits. This increase could of mercury into the surrounding ecosystem due to be coming from contaminated deposits that are improper protections at the informal mining sites. slowly moving downstream in the wetland, showing Mercury is a strong neurotoxin and can bioaccumulate the ability of the wetland to store and attenuate the in tissue, meaning that a larger organism can retain movement of these heavy metal deposits (Subalusky and build up mercury that comes from the smaller et al., 2019). A point of concern is that mercury in organisms it consumes. In the Tigithe River, there low oxygen environments (like wetland soils) has is concern that fish in the system may contain high the potential to convert to methyl mercury, the most amounts of mercury that could be dangerous for toxic form of mercury for living organisms. human consumption. In addition, there have been
240 Studies on fish in the Tigithe River and Lake Victoria hippopotamuses. This is particularly important have also been conducted to see if there is an impact at the Kogatende site inside SENAPA as there are on aquatic species and how far that impact may travel large populations of these species in this area. These downstream. They found that, in general, the fish species are not a specific indicator for this site since tested in the Tigithe River had only slightly higher they are not true aquatic species, but they have levels of heavy metals to those found in Lake Victoria, been considered as part of the process. The low flow indicating that the mining activities were not having recommendations for both maintenance years and a significant impact on aquatic species in that trophic drought years consider keeping pools deep enough level (Machiwa, 2003; Mataba et al., 2016). Another to provide habitat for both species, which is often a study in northwestern Tanzania found elevated very stressful time period. Since both of these species levels of mercury in individual fish at sites highly are mobile, it is expected that they move up and contaminated by artisanal mining, but that these downstream to find pools of an appropriate size. levels were not seen in the general fish population in Downstream of Kogatende, there is less emphasis other parts of the system and downstream in Lake on these species. This is because there are reduced Tanganyika (Taylor et al., 2005). The levels of heavy numbers due to human encroachment on the metals found in the Tigithe River were below the mainstem sites and the fact that these species don’t Provisional Tolerable Weekly Intake level when fish live in the tributaries (Tobora, Somoche, and Tigithe). is consumed in average amounts, but there may be There have been some sightings of crocodiles and concerns for people who consume large amounts of hippopotamuses in the wetland by local communities, fish daily as well as at-risk populations (Mataba et al., but there is enough water in that habitat for their 2016). needs and it is not a major concern at that site. While water quality does play an important role when 6.5 Integration into Existing deciding environmental flow recommendations, and Future Water Resources the focus is on parameters that may be impacted Management by an extreme or persistent change in flow, such as The EFA is one piece of many integrated water temperature, dissolved oxygen, and nutrients. Heavy resources management plans and activities in the metals occur naturally in the earth’s crust and are Lower MRB in Tanzania, the entire MRB including found in water and sediments at trace levels, which Kenya, and the East Africa region at large. In the do not pose a health concern to aquatic organisms. Lower MRB, there are multiple management plans Elevated levels of heavy metals may pose a threat to already in place in various parts of the catchment that aquatic ecosystems and local communities, but are include environmental flows directly or indirectly. most likely due to non-flow related human activities The Mara Wetlands Integrated Management Plan like mining or other industrial processes. However, (URT, 2018b), which is being carried out under environmental flows are not an appropriate the Mara Regional Commissioner’s Office, has management tool for elevated levels of heavy metal environmental flows listed as Activity 1.7 under the pollutants and should not be used as a dilution Land Use and Wetland Management Programme method for human-introduced contaminants. as a way to regulate water abstractions. Six WUAs The management of these pollutants should be have been created in the Lower MRB, two of which conducted separately from environmental flows (Somoche and Tobora WUAs) have created and and be controlled through other activities, such as approved subcatchment management plans. The being included in local management plans and/or LVBWB is also in the process of developing their the development of pollution prevention and control Integrated Water Resources Management and plans for specific activities. Development Plan for the Lake Victoria Basin, in which determining and implementing the reserve is 6.4.3 Wildlife likely to play a foundational role in planning for future The Serengeti ecosystem hosts the world’s largest water resources development. A previous project overland migration, including 1,300,000 wildebeest, funded by USAID project (Planning for Resilience in 200,000 zebras, 350,000 gazelles, and 12,000 East Africa through Policy, Adaptation, Research and elands cross the Mara River each year between Economic Development, or PREPARED) developed June and November (Hopcraft, 2010; Tanzania an economic valuation of biodiversity and ecosystem Tourism Board, 2012; Subalusky et al., 2017). Most services in the Mara Wetlands (USAID, 2016) and a of these animals cross through the Lower MRB as conservation investment plan for the Mara Wetland they migrate between SENAPA in Tanzania and the (URT, 2017), both of which support the protection Maasai Mara National Reserve in Kenya and consume and sustainable use of the Mara Wetland. surface water along their journey. The Mara River is a critical water resource during this journey since it For many years, activities have been conducted in is the only perennial water source in the area. The the both Kenyan and Tanzanian sides of the basin to water required for the migration is important, but is protect aquatic ecosystems and biodiversity, ensure not directly related to protecting aquatic ecosystems the responsible use of water resources, and prepare and hence is not included in the environmental flow for impacts from climate change, which all have recommendations. This consumptive water use is links to environmental flows. The 2010 Biodiversity considered as a water demand in the parallel WAP Strategy and Action Plan for Sustainable Management process. There is also an important consideration of the Mara River Basin (EAC and WWF-ESARPO, when it comes to animals which use the Mara River 2010) calls for establishing and implementing the as their primary habitat, such as crocodiles and reserve as part of its objectives for aquatic habitats.
241 Between 2014 and 2018, the MaMaSe project worked and sustainable use of wetlands. At the Nile Basin in Kenya on sustainable resources management, scale, NBI considers “establishment of thresholds for including developing water abstraction reports for sustainable flow requirements” in wetlands as one each subcatchment, a draft environmental flow their priority outputs in their Wetland Management assessment, and a preliminary WAP calculations for Strategy (NBI, 2013), which is directly related to the the MRB located on the Kenya side. The SWP Mara motivation for this effort. River Basin Activity under USAID is supporting a series of efforts to develop a transboundary WAP Looking towards the future, these efforts create a between Kenya and Tanzania, of which this report is strong foundation for further financial investment a critical component. This report was also developed in the basin, particularly around climate change and to align with procedures followed under the MaMaSe sustainable management of water resources. With project so the outcomes could be readily combined various sources of global and regional funds available as part of the transboundary process. WWF has for climate change mitigation and adaptation, also developed a Climate Change Vulnerability funders are looking for “bankable” projects. These and Adaptation Assessment for the Greater Mara projects should define specific issues related to Ecosystem, covering both the Maasai Mara National climate change in the basin, clearly articulate how the Reserve in Kenya and SENAPA in Tanzania (WWF- proposed actions are going to mitigate or adaptive Kenya, 2019), in which environmental flows is listed to changes due to climate change, identify a series as an important activity for increasing resilience to of ready-to-implement projects, describe how they climate change. will reduce potential risks, and align with existing regional management plans and climate strategies Regionally, there has been many similar efforts (World Bank Group, 2019). Transboundary basins focused on the Lake Victoria Basin, the East African also present an additional challenge of needing Community, as well as the larger Nile River Basin. a river basin organization to bring together the Under PREPARED, a climate change adaptation individual countries and act as the overall project strategy and action plan for 2018-2023 for the Lake manager. The MRB already meets many of these Victoria Basin was developed by the Lake Victoria requirements through the outputs from the Basin Commission that outlined activities to protect efforts listed above, including preparation of the water and aquatic ecosystems (LVBC, 2018). The Conservation Investment Plan for Mara Wetlands, East African Community also developed their and could function through the existing river basin own climate change master plan for 2011 – 2031 organization, the Lake Victoria Basin Commission. (EAC, 2011), where adaption measures for water This EFA acts as a project preparation study, security include the promotion of integrated water providing technical backing to future “bankable” resources management, protection of watersheds,
242 243 7. REFERENCES projects in the MRB, particularly those related to mitigating impacts on aquatic ecosystems and sustainable water resources management. Almås, Å. R. and Manoko, M. L. K. (2012) ‘Trace Element Concentrations in Soil, Sediments, and Waters in the Vicinity of Geita Gold Mines and North Mara Gold Mines in Northwest Tanzania’, Soil and Sediment Contamination, 21(2), pp. 135–159. doi: 10.1080/15320383.2012.649372. Bitala, M. F., Kweyunga, C. and Manoko, M. L. (2009) Levels of Heavy Metals and Cyanide in Soil, Sediment and Water from the Vicinity of North Mara Gold Mine in Tarime District, Tanzania. Boney, A. (1989) Phytoplankton: Great Britain. London: Chapman and Hall. Bora, M. and Goswami, D. C. (2017) ‘Water quality assessment in terms of water quality index (WQI): case study of the Kolong River, Assam, India’, Applied Water Science. Springer Berlin Heidelberg, 7(6), pp. 3125–3135. doi: 10.1007/s13201-016-0451-y. Braak, C. J. F. ter and Smilauer, P. (2002) CANOCO Reference Manual and CanoDraw for Windows User’s Guide: Software for Canonical Community Ordination (version 4.5). 33, , : www. canoco.com. Available at: http://edepot.wur.nl/405659. Brisbane Declaration (2007) ‘The Brisbane Declaration: Environmental Flows are Essential for Freshwater Ecosystem Health and Human Well-Being.’, in Declaration of the 10th International Riversymposium and International Environmental Flows Conference. Brisbane, Australia. 3-6 September 2007. Camargo, J. A., Alonso, A. and De La Puente, M. (2004) ‘Multimetric assessment of nutrient enrichment in impounded rivers based on benthic macroinvertebrates’, Environmental Monitoring and Assessment, 96(1–3), pp. 233–249. doi: 10.1023/B:EMAS.0000031730.78630.75. CDM Smith (2016) Environmental Flows in Rufiji River Basin Assessed from the Perspective of Planned Development in Kilombero and Lower Rufiji Sub-Basins. CDM Smith (2018) Draft. Reserve Monitoring Program and Adaptive Management Plan, Rufiji Basin, Tanzania. Connolly, N. M., Crossland, M. R. and Pearson, R. G. (2004) ‘Effect of low dissolved oxygen on survival, emergence, and drift of tropical stream macroinvertebrates’, Journal of the North American Benthological Society, 23(2), pp. 251–270. doi: 10.1899/0887-3593(2004)023<0251:eol doo>2.0.co;2. Dallas, H. (2008) ‘Water temperature and riverine ecosystems: An overview of knowledge and approaches for assessing biotic responses, with special reference to South Africa’, 34(3), pp. 393– 404. Dessu, S. B. and Melesse, A. M. (2012) ‘Impact and uncertainties of climate change on the hydrology of the Mara River basin, Kenya/Tanzania’, Hydrological Processes, 27(20), pp. 2973– 2986. doi: 10.1002/hyp.9434. EAC, (East African Community) (2011) East African Community Climate Change Master Plan 2011- 2031. Available at: https://www.eac.int/documents/category/environment-and-natural-resources. EAC, (East African Community) and WWF-ESARPO, (WWF Eastern and Southern African Regional Programme Office) (2010) Biodiversity Strategy and Action Plan for Sustainable Management of the Mara River Basin. Available at: http://d2ouvy59p0dg6k.cloudfront.net/downloads/biodiverstiy_ strategy_action_plan_mara_1.pdf. Emlid (2019) RTK GNSS receiver, https://emlid.com/reachrs/; Access Date: 2019-07-23. GLOWS-FIU, (Global Water for Sustainability Program) (2012) Environmental Flow Recommendations for Resere Flows in the Mara River, Kenya and Tanzania. GLOWS-FIU, (Global Water for Sustainability Program) and WWF-ESARPO, (WWF Eastern and Southern African Regional Programme Office) (2007) Water Quality Baseline Assessment Report : Mara River Basin , Kenya-Tanzania. Hayes, J. W. and Young, R. G. (2001) ‘Effects of Low Flow on Trout and Salmon in Relation to the Regional Water Plan : Otago Prepared for Fish and Game New Zealand Effects of Low Flow on Trout 244 and Salmon in Relation to the Regional Water Plan : Otago Prepared for Fish and Game New Zealand by’, Cawthron Report No. 615, (615). Hopcraft, J. G. C. (2010) Ecological implications of food and predation risk for herbivores in the Serengeti. Ikingura, J. R. et al. (2006) ‘Environmental assessment of mercury dispersion, transformation and bioavailability in the Lake Victoria Goldfields, Tanzania’, Journal of Environmental Management, 81(2), pp. 167–173. doi: 10.1016/j.jenvman.2005.09.026. Jackson, J. (1997) ‘State of Habitat Availability and Quality in Inland Waters’, Australia: State of the environment technical papers series (Inland waters). Kihampa C and Wenaty A (2013) ‘Impact of Mining and Farming activities on Water and Sediment Quality of the Mara river basin, Tanzania’, Research Journal of Chemical Sciences, 3(7), pp. 15–24. King, J., Tharme, R. and Villiers, M. De (2008) Environmental flow assessments for rivers: manual for the Building Block Methodology. Available at: http://swrcb2.swrcb.ca.gov/waterrights/water_issues/programs/ bay_delta/deltaflow/docs/exhibits/swrcb/swrcb_king2008.pdf. Kleynhans, C. J. (1999) ‘The development of a fish index to assess the biological integrity of South African rivers’, Water SA, 25(3), pp. 265–278. Koehn, J., O’Connor, W. G. and Research, A. R. I. for E. (1990) ‘Biological information for management of native freshwater fish in Victoria’. Melbourne, Vic: Dept. of Conservation and Environment, Freshwater Fish Management Branch, Arthur Rylah Institute for Environmental Research. LVBC, (Lake Victoria Basin Commission of the East African Community) (2018) ‘Lake Victoria Basin: Climate Change Adaptation Strategy and Action Plan 2018-2023’, p. 76. Available at: https://www.climatelinks.org/ sites/default/files/asset/document/2018_USAID-PREPARED_Lake-Victoria-Basin-Adaptation-Strategy- Action-Plan.pdf. Machiwa, J. (2003) ‘Metal concentrations in sediment and fish of Lake Victoria near and away from catchments with gold mining acctivities’, Tanzania Journal of Science. doi: 10.4314/tjs.v29i2.18377. MaMaSe, (Mau Mara Serengeti Sustainable Water Initiative) (2017) Preliminary Water Resources Assessment of the Mara River Basin (J. Wenninger, R. Venneker); Internal MaMaSe report, Sub-Result area 1, January 2017. MaMaSe, (Mau Mara Serengeti Sustainable Water Initiative) (2019) ‘Land-use map of the Mara River Basin, Publication Date: April 3, 2019, 9:23 a.m., http://maps.mamase.org/maps/626; Access Date: 2019-07-23’. Mango, L. M. et al. (2011) ‘Land use and climate change impacts on the hydrology of the upper Mara River Basin, Kenya: Results of a modeling study to support better resource management’, Hydrology and Earth System Sciences, 15(7), pp. 2245–2258. doi: 10.5194/hess-15-2245-2011. Mataba, G. R. et al. (2016) ‘Distribution of trace elements in the aquatic ecosystem of the Thigithe river and the fish Labeo victorianus in Tanzania and possible risks for human consumption’, Science of the Total Environment, 547, pp. 48–59. doi: 10.1016/j.scitotenv.2015.12.123. Mganga, N., Manoko, M. L. K. and Rulangaranga, Z. K. (2011) ‘Classification of Plants According to Their Heavy Metal Content around North Mara Gold Mine, Tanzania: Implication for Phytoremediation’, Tanzania Journal of Science. MoWI, M. of W. and I. U. R. of T. (2009) ‘Design Manual for Water Supply and Waste Water Disposal. Third Edition. Chapter 4.’ NBI, (Nile Basin Initiative) (2013) Wetland Management Strategy. Available at: http://nileis.nilebasin.org/ system/files/12 11 13 wetland management strategy.pdf. NBI, (Nile Basin Initiative) (2015) Mara River Basin Management Project. Available at: http://nelsap. nilebasin.org/attachments/article/34/Mara Fact Sheet May2015.pdf. NBI, (Nile Basin Initiative) (2016a) Preparation of NBI Guidance Document on Environmental Flows: Nile E-flows Framework Technical Implementation Manual. Prepared by HYDROC GmbH on behalf of the Nile Basin Initiative and Deutsche Gesellschaft für Internationale Zusammenarbeit. Contract No. 8. NBI, (Nile Basin Initiative) (2016b) Strategy for Management of Environmental Flows in the Nile Basin.
245 NBS, N. B. of S. (2012) Population and Housing Census: Population Distribution by Administrative Areas. United Republic of Tanzania. Republic of Kenya (2012) Water Act. The Republic of Kenya: Ministry of Water and Irrigation. Richter, B. D. et al. (1996) ‘A Method for Assessing Hydrologic Alteration within Ecosystems’, Conservation Biology, 10(4), pp. 1163–1174. Rowntree, K. M. and Wadeson, R. A. (1999) A hierarchical geomorphological model for the classification of selected South African rivers. Water Research Commission Report 497/1/99. Pretoria, South Africa. Roy, T. et al. (2018) ‘Assessing hydrological impacts of short-term climate change in the Mara River basin of East Africa’, Journal of Hydrology, 566(March), pp. 818–829. doi: 10.1016/j.jhydrol.2018.08.051. Scott, D. F. and Lesch, W. (1997) ‘Streamflow responses to afforestation with Eucalyptus grundis and Pinus putulu and to felling in the Mokobulaan experimental catchments, South Africa’, Journal of Hydrology, 199, pp. 360–377. Subalusky, A. A. et al. (2019) Reading the Historical Record of the Mara Using Sediment Cores from the Mara Wetland. Final Report. Subalusky, A. L. et al. (2017) ‘Annual mass drownings of the Serengeti wildebeest migration influence nutrient cycling and storage in the Mara River’, 114(29), pp. 7647–7652. doi: 10.1073/pnas.1614778114.
Subalusky, A. L. et al. (2018) ‘Organic matter and nutrient inputs from large wildlife influence ecosystem function in the Mara River, Africa’, Ecology, 99(11), pp. 2558–2574. doi: 10.1002/ecy.2509. SWP, (Sustainable Water Partnership) (2018) Assessment of the hydrological monitoring network in the Mara river basin - Tanzania (J. Wenninger, R. Venneker); Internal SWP report, IHE-Delft Task 4; Subtask 4.1.3, September 2018. SWP, (Sustainable Water Partnership) (2019) Water Availability Assessment - Lower Mara River Basin, Tanzania, (J. Wenninger, R. Venneker); Technical Annex, IHE-Delft Task 4; Subtask 4.1.3, Internal Report. Tanzania Tourism Board (2012) The Great Serengeti Migration, www.tanzaniatourism.go.tz. Available at: tanzaniatourism.go.tz/en/highlights/view/the-great-serengeti-migration (Accessed: 9 September 2019). Taylor, H. et al. (2005) ‘Environmental assessment of mercury contamination from the Rwamagasa artisanal gold mining centre, Geita District, Tanzania’, Science of the Total Environment, 343(1–3), pp. 111–133. doi: 10.1016/j.scitotenv.2004.09.042. The Nature Conservancy (2009) ‘Indicators of Hydrologic Alteration Version 7.1 User’s Manual’, p. 76. UNEP/UNU-EHS (2016) International Water Quality Guidelines for Ecosystems (IWQGES). How to develop guidelines for healthy freshwater ecosystems. A policy oriented approach. Available at: http://web.unep.org/ sites/default/files/Documents/20160315_iwqges_pd_final.pdf. URT, (United Republic of Tanzania) (1997) ‘National Environmental Policy’, pp. 1–35. URT, (United Republic of Tanzania) (2002) ‘National Water Policy - July 2002’, p. 109. doi: 10.1093/chemse/ bjt099. URT, (United Republic of Tanzania) (2004) ‘Environmental Management Act’. URT, (United Republic of Tanzania) (2006) Water Sector Development Programme 2006 - 2025. URT, (United Republic of Tanzania) (2008) National Water Development Strategy, 2006 to 2015. URT, (United Republic of Tanzania) (2009) ‘Water Resources Management Act, 2009’, p. 73. URT, (United Republic of Tanzania) (2014) Second National Communication to the United Nations Framework Convention on Climate Change. Available at: https://unfccc.int/sites/default/files/resource/tzanc2.pdf. URT, (United Republic of Tanzania) (2016) Environmental Water Requirements Assessment Guidelines for Tanzania. Final Draft. URT, (United Republic of Tanzania) (2017) Conservation Investment Plan for Mara Wetlands, United Republic
246 Of Tanzania. Available at: https://www.climatelinks.org/sites/default/files/asset/document/2018_USAID- PREPARED_Conservation-Plan-for-Mara-Wetlands.pdf. URT, (United Republic of Tanzania) (2018a) Draft Tanzania Guidelines for Water Allocation Planning. URT, (United Republic of Tanzania) (2018b) Mara Wetlands Integrated Management Plan 2018 - 2022. URT, (United Republic of Tanzania) and Republic of Kenya (2015) Memorandum of Understanding between the Government of the Republic of Kenya and the Government of the United Republic of Tanzania for Joint Water Resources Management of the Transboundary Mara River Basin. USAID (2018) A River Classification System for the Rufiji River Basin and Other Major Basins of Tanzania. Technical Assistance to Support the Development of Irrigation and Rural Roads Infrastructure Project (IRRIP2). Draft. USAID (2019) Vulnerability and Adaptation in the Mara River Basin. USAID, (United States Agency for Interational Development) (2016) Economic Valuation of Biodiversity and Ecosystem Services in the Mara Wetlands, United Republic of Tanzania. USAID, (United States Agency for Interational Development) (2017) Vulnerability, Impact and Adaptation Assessment in the East Africa Region. Chapter 9: Water, Aqautic Ecosystems, and Water Supply Infrastructure - Future Impacts from Climate Chnage.
Venneker, R. (2011) ‘DScreen User Manual (Version 1.0 $Exp), UNESCO-IHE, Delft, The Netherlands’. World Bank Group (2019) Financing Climate Change Adaptation in Transboundary Basins: Preparing Bankable. doi: 10.1596/31224. WRMA, (Water Resource Management Authority) (2009) Final Water Allocation Guidelines.
247 248 7. ANNEXES
WWF-Kenya (2019) Climate Change Vulnerability and Adaptation Assessment for the Greater Mara Ecosystem Climate Change Vulnerability and Adaptation Assessment for the Greater Mara Ecosystem. Annex A : Stakeholder Input on Determining Resource Quality Objectives for the Lower Mara River, Workshop Report Annex B: Starter Document for Socioeconomics Annex C: Starter Document for Hydrology Annex D: Starter Document for Hydraulics Annex E: Starter Document for Geomorphology Annex F: Starter Document for Riparian Vegetation Annex G: Starter Document for Fish Annex H: Starter Document for Macroinvertebrates Annex I: Starter Document for Water Quality
249 250 251 © Copyright 2020 Nile Basin Initiative (NBI)
Lower Mara Environmental Flow Assessment: Resource Quality Objectives and Reserve Assessment Report. Prepared by IHE Delft Institute for Water Education on behalf of the Nile Equatorial Lakes Subsidiary Action Program Coordination Unit (NELSAP-CU/NBI) “ This activity was done with support from GIZ, on behalf of the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) under the International Climate Initiative.” Project number 14.9029.1 ONE RIVER ONE PEOPLE ONE VISION
Nile Equatorial Lakes Subsidiary Action Program (NELSAP-CU) Kigali City Tower, 5th Floor, P. O. Box 6759, KN 81 Street Kigali, Rwanda Tel: (250) 788 307 334 Twitter: NelsapCu, Facebook: NelsapCu/ Email: [email protected] www.nelsap.nilebasin.org